#!./perl
-print "1..61\n";
+BEGIN {
+ chdir 't' if -d 't';
+ @INC = qw(. ../lib);
+ require 'test.pl';
+}
+
+use strict qw(refs subs);
+
+plan(235);
# Test glob operations.
-$bar = "ok 1\n";
-$foo = "ok 2\n";
+$bar = "one";
+$foo = "two";
{
local(*foo) = *bar;
- print $foo;
+ is($foo, 'one');
}
-print $foo;
+is ($foo, 'two');
-$baz = "ok 3\n";
-$foo = "ok 4\n";
+$baz = "three";
+$foo = "four";
{
local(*foo) = 'baz';
- print $foo;
+ is ($foo, 'three');
}
-print $foo;
+is ($foo, 'four');
-$foo = "ok 6\n";
+$foo = "global";
{
local(*foo);
- print $foo;
- $foo = "ok 5\n";
- print $foo;
+ is ($foo, undef);
+ $foo = "local";
+ is ($foo, 'local');
}
-print $foo;
+is ($foo, 'global');
+{
+ no strict 'refs';
# Test fake references.
-$baz = "ok 7\n";
-$bar = 'baz';
-$foo = 'bar';
-print $$$foo;
+ $baz = "valid";
+ $bar = 'baz';
+ $foo = 'bar';
+ is ($$$foo, 'valid');
+}
# Test real references.
$FOO = \$BAR;
$BAR = \$BAZ;
-$BAZ = "ok 8\n";
-print $$$FOO;
+$BAZ = "hit";
+is ($$$FOO, 'hit');
# Test references to real arrays.
-@ary = (9,10,11,12);
+my $test = curr_test();
+@ary = ($test,$test+1,$test+2,$test+3);
$ref[0] = \@a;
$ref[1] = \@b;
$ref[2] = \@c;
print @a;
print ${$ref[1]}[0];
print @{$ref[2]}[0];
-print @{'d'};
+{
+ no strict 'refs';
+ print @{'d'};
+}
+curr_test($test+4);
# Test references to references.
$refref = \\$x;
-$x = "ok 13\n";
-print $$$refref;
+$x = "Good";
+is ($$$refref, 'Good');
# Test nested anonymous lists.
$ref = [[],2,[3,4,5,]];
-print scalar @$ref == 3 ? "ok 14\n" : "not ok 14\n";
-print $$ref[1] == 2 ? "ok 15\n" : "not ok 15\n";
-print ${$$ref[2]}[2] == 5 ? "ok 16\n" : "not ok 16\n";
-print scalar @{$$ref[0]} == 0 ? "ok 17\n" : "not ok 17\n";
+is (scalar @$ref, 3);
+is ($$ref[1], 2);
+is (${$$ref[2]}[2], 5);
+is (scalar @{$$ref[0]}, 0);
-print $ref->[1] == 2 ? "ok 18\n" : "not ok 18\n";
-print $ref->[2]->[0] == 3 ? "ok 19\n" : "not ok 19\n";
+is ($ref->[1], 2);
+is ($ref->[2]->[0], 3);
# Test references to hashes of references.
$refref = \%whatever;
$refref->{"key"} = $ref;
-print $refref->{"key"}->[2]->[0] == 3 ? "ok 20\n" : "not ok 20\n";
+is ($refref->{"key"}->[2]->[0], 3);
# Test to see if anonymous subarrays spring into existence.
$spring[5]->[0] = 123;
$spring[5]->[1] = 456;
push(@{$spring[5]}, 789);
-print join(':',@{$spring[5]}) eq "123:456:789" ? "ok 21\n" : "not ok 21\n";
+is (join(':',@{$spring[5]}), "123:456:789");
# Test to see if anonymous subhashes spring into existence.
@{$spring2{"foo"}} = (1,2,3);
$spring2{"foo"}->[3] = 4;
-print join(':',@{$spring2{"foo"}}) eq "1:2:3:4" ? "ok 22\n" : "not ok 22\n";
+is (join(':',@{$spring2{"foo"}}), "1:2:3:4");
# Test references to subroutines.
-sub mysub { print "ok 23\n" }
-$subref = \&mysub;
-&$subref;
+{
+ my $called;
+ sub mysub { $called++; }
+ $subref = \&mysub;
+ &$subref;
+ is ($called, 1);
+}
+is ref eval {\&{""}}, "CODE", 'reference to &{""} [perl #94476]';
+
+# Test references to return values of operators (TARGs/PADTMPs)
+{
+ my @refs;
+ for("a", "b") {
+ push @refs, \"$_"
+ }
+ is join(" ", map $$_, @refs), "a b", 'refgen+PADTMP';
+}
$subrefref = \\&mysub2;
-$$subrefref->("ok 24\n");
-sub mysub2 { print shift }
+is ($$subrefref->("GOOD"), "good");
+sub mysub2 { lc shift }
+
+# Test REGEXP assignment
+
+SKIP: {
+ skip_if_miniperl("no dynamic loading on miniperl, so can't load re", 5);
+ require re;
+ my $x = qr/x/;
+ my $str = "$x"; # regex stringification may change
+
+ my $y = $$x;
+ is ($y, $str, "bare REGEXP stringifies correctly");
+ ok (eval { "x" =~ $y }, "bare REGEXP matches correctly");
+
+ my $z = \$y;
+ ok (re::is_regexp($z), "new ref to REXEXP passes is_regexp");
+ is ($z, $str, "new ref to REGEXP stringifies correctly");
+ ok (eval { "x" =~ $z }, "new ref to REGEXP matches correctly");
+}
+{
+ my ($x, $str);
+ {
+ my $y = qr/x/;
+ $str = "$y";
+ $x = $$y;
+ }
+ is ($x, $str, "REGEXP keeps a ref to its mother_re");
+ ok (eval { "x" =~ $x }, "REGEXP with mother_re still matches");
+}
# Test the ref operator.
-print ref $subref eq CODE ? "ok 25\n" : "not ok 25\n";
-print ref $ref eq ARRAY ? "ok 26\n" : "not ok 26\n";
-print ref $refref eq HASH ? "ok 27\n" : "not ok 27\n";
+sub PVBM () { 'foo' }
+{ my $dummy = index 'foo', PVBM }
+
+my $pviv = 1; "$pviv";
+my $pvnv = 1.0; "$pvnv";
+my $x;
+
+# we don't test
+# tied lvalue => SCALAR, as we haven't tested tie yet
+# BIND, 'cos we can't create them yet
+# REGEXP, 'cos that requires overload or Scalar::Util
+
+for (
+ [ 'undef', SCALAR => \undef ],
+ [ 'constant IV', SCALAR => \1 ],
+ [ 'constant NV', SCALAR => \1.0 ],
+ [ 'constant PV', SCALAR => \'f' ],
+ [ 'scalar', SCALAR => \$x ],
+ [ 'PVIV', SCALAR => \$pviv ],
+ [ 'PVNV', SCALAR => \$pvnv ],
+ [ 'PVMG', SCALAR => \$0 ],
+ [ 'PVBM', SCALAR => \PVBM ],
+ [ 'scalar @array', SCALAR => \scalar @array ],
+ [ 'scalar %hash', SCALAR => \scalar %hash ],
+ [ 'vstring', VSTRING => \v1 ],
+ [ 'ref', REF => \\1 ],
+ [ 'substr lvalue', LVALUE => \substr($x, 0, 0) ],
+ [ 'pos lvalue', LVALUE => \pos ],
+ [ 'vec lvalue', LVALUE => \vec($x,0,1) ],
+ [ 'named array', ARRAY => \@ary ],
+ [ 'anon array', ARRAY => [ 1 ] ],
+ [ 'named hash', HASH => \%whatever ],
+ [ 'anon hash', HASH => { a => 1 } ],
+ [ 'named sub', CODE => \&mysub, ],
+ [ 'anon sub', CODE => sub { 1; } ],
+ [ 'glob', GLOB => \*foo ],
+ [ 'format', FORMAT => *STDERR{FORMAT} ],
+) {
+ my ($desc, $type, $ref) = @$_;
+ is (ref $ref, $type, "ref() for ref to $desc");
+ like ("$ref", qr/^$type\(0x[0-9a-f]+\)$/, "stringify for ref to $desc");
+}
+
+is (ref *STDOUT{IO}, 'IO::File', 'IO refs are blessed into IO::File');
+like (*STDOUT{IO}, qr/^IO::File=IO\(0x[0-9a-f]+\)$/,
+ 'stringify for IO refs');
+
+{ # Test re-use of ref's TARG [perl #101738]
+ my $obj = bless [], '____';
+ my $uniobj = bless [], chr 256;
+ my $get_ref = sub { ref shift };
+ my $dummy = &$get_ref($uniobj);
+ $dummy = &$get_ref($obj);
+ ok exists { ____ => undef }->{$dummy}, 'ref sets UTF8 flag correctly';
+}
# Test anonymous hash syntax.
$anonhash = {};
-print ref $anonhash eq HASH ? "ok 28\n" : "not ok 28\n";
-$anonhash2 = {FOO => BAR, ABC => XYZ,};
-print join('', sort values %$anonhash2) eq BARXYZ ? "ok 29\n" : "not ok 29\n";
+is (ref $anonhash, 'HASH');
+$anonhash2 = {FOO => 'BAR', ABC => 'XYZ',};
+is (join('', sort values %$anonhash2), 'BARXYZ');
# Test bless operator.
package MYHASH;
$object = bless $main'anonhash2;
-print ref $object eq MYHASH ? "ok 30\n" : "not ok 30\n";
-print $object->{ABC} eq XYZ ? "ok 31\n" : "not ok 31\n";
+main::is (ref $object, 'MYHASH');
+main::is ($object->{ABC}, 'XYZ');
$object2 = bless {};
-print ref $object2 eq MYHASH ? "ok 32\n" : "not ok 32\n";
+main::is (ref $object2, 'MYHASH');
# Test ordinary call on object method.
-&mymethod($object,33);
+&mymethod($object,"argument");
sub mymethod {
local($THIS, @ARGS) = @_;
die 'Got a "' . ref($THIS). '" instead of a MYHASH'
- unless ref $THIS eq MYHASH;
- print $THIS->{FOO} eq BAR ? "ok $ARGS[0]\n" : "not ok $ARGS[0]\n";
+ unless ref $THIS eq 'MYHASH';
+ main::is ($ARGS[0], "argument");
+ main::is ($THIS->{FOO}, 'BAR');
}
# Test automatic destructor call.
-$string = "not ok 34\n";
+$string = "bad";
$object = "foo";
-$string = "ok 34\n";
+$string = "good";
$main'anonhash2 = "foo";
$string = "";
DESTROY {
return unless $string;
- print $string;
+ main::is ($string, 'good');
# Test that the object has not already been "cursed".
- print ref shift ne HASH ? "ok 35\n" : "not ok 35\n";
+ main::isnt (ref shift, 'HASH');
}
# Now test inheritance of methods.
package OBJ;
-@ISA = (BASEOBJ);
+@ISA = ('BASEOBJ');
-$main'object = bless {FOO => foo, BAR => bar};
+$main'object = bless {FOO => 'foo', BAR => 'bar'};
package main;
# Test arrow-style method invocation.
-print $object->doit("BAR") eq bar ? "ok 36\n" : "not ok 36\n";
+is ($object->doit("BAR"), 'bar');
# Test indirect-object-style method invocation.
$foo = doit $object "FOO";
-print $foo eq foo ? "ok 37\n" : "not ok 37\n";
+main::is ($foo, 'foo');
sub BASEOBJ'doit {
local $ref = shift;
- die "Not an OBJ" unless ref $ref eq OBJ;
+ die "Not an OBJ" unless ref $ref eq 'OBJ';
$ref->{shift()};
}
@ISA = 'LASTCHANCE';
package LASTCHANCE;
-sub foo { print $_[1] }
+sub foo { main::is ($_[1], 'works') }
package WHATEVER;
-foo WHATEVER "ok 38\n";
+foo WHATEVER "works";
#
# test the \(@foo) construct
#
package main;
-@foo = (1,2,3);
+@foo = \(1..3);
@bar = \(@foo);
@baz = \(1,@foo,@bar);
-print @bar == 3 ? "ok 39\n" : "not ok 39\n";
-print grep(ref($_), @bar) == 3 ? "ok 40\n" : "not ok 40\n";
-print @baz == 3 ? "ok 41\n" : "not ok 41\n";
+is (scalar (@bar), 3);
+is (scalar grep(ref($_), @bar), 3);
+is (scalar (@baz), 3);
-my(@fuu) = (1,2,3);
+my(@fuu) = \(1..2,3);
my(@baa) = \(@fuu);
my(@bzz) = \(1,@fuu,@baa);
-print @baa == 3 ? "ok 42\n" : "not ok 42\n";
-print grep(ref($_), @baa) == 3 ? "ok 43\n" : "not ok 43\n";
-print @bzz == 3 ? "ok 44\n" : "not ok 44\n";
+is (scalar (@baa), 3);
+is (scalar grep(ref($_), @baa), 3);
+is (scalar (@bzz), 3);
-# test for proper destruction of lexical objects
+# also, it can't be an lvalue
+eval '\\($x, $y) = (1, 2);';
+like ($@, qr/Can\'t modify.*ref.*in.*assignment/);
-sub larry::DESTROY { print "# larry\nok 45\n"; }
-sub curly::DESTROY { print "# curly\nok 46\n"; }
-sub moe::DESTROY { print "# moe\nok 47\n"; }
+# test for proper destruction of lexical objects
+$test = curr_test();
+sub larry::DESTROY { print "# larry\nok $test\n"; }
+sub curly::DESTROY { print "# curly\nok ", $test + 1, "\n"; }
+sub moe::DESTROY { print "# moe\nok ", $test + 2, "\n"; }
{
my ($joe, @curly, %larry);
}
print "# left block\n";
+curr_test($test + 3);
# another glob test
-$foo = "not ok 48";
+
+$foo = "garbage";
{ local(*bar) = "foo" }
-$bar = "ok 48";
+$bar = "glob 3";
local(*bar) = *bar;
-print "$bar\n";
+is ($bar, "glob 3");
-$var = "ok 49";
+$var = "glob 4";
$_ = \$var;
-print $$_,"\n";
+is ($$_, 'glob 4');
-# test if reblessing during destruction results in more destruction
+# test if reblessing during destruction results in more destruction
+$test = curr_test();
{
package A;
sub new { bless {}, shift }
- DESTROY { print "# destroying 'A'\nok 51\n" }
+ DESTROY { print "# destroying 'A'\nok ", $test + 1, "\n" }
package _B;
sub new { bless {}, shift }
- DESTROY { print "# destroying '_B'\nok 50\n"; bless shift, 'A' }
+ DESTROY { print "# destroying '_B'\nok $test\n"; bless shift, 'A' }
package main;
my $b = _B->new;
}
+curr_test($test + 2);
# test if $_[0] is properly protected in DESTROY()
{
+ my $test = curr_test();
my $i = 0;
local $SIG{'__DIE__'} = sub {
my $m = shift;
if ($i++ > 4) {
- print "# infinite recursion, bailing\nnot ok 52\n";
+ print "# infinite recursion, bailing\nnot ok $test\n";
exit 1;
}
- print "# $m";
- if ($m =~ /^Modification of a read-only/) { print "ok 52\n" }
+ like ($m, qr/^Modification of a read-only/);
};
package C;
sub new { bless {}, shift }
print "# good, didn't recurse\n";
}
-# test if refgen behaves with autoviv magic
+# test that DESTROY is called on all objects during global destruction,
+# even those without hard references [perl #36347]
+
+is(
+ runperl(
+ stderr => 1, prog => 'sub DESTROY { print qq-aaa\n- } bless \$a[0]'
+ ),
+ "aaa\n", 'DESTROY called on array elem'
+);
+is(
+ runperl(
+ stderr => 1,
+ prog => '{ bless \my@x; *a=sub{@x}}sub DESTROY { print qq-aaa\n- }'
+ ),
+ "aaa\n",
+ 'DESTROY called on closure variable'
+);
+
+# But cursing objects must not result in double frees
+# This caused "Attempt to free unreferenced scalar" in 5.16.
+fresh_perl_is(
+ 'bless \%foo::, bar::; bless \%bar::, foo::; print "ok\n"', "ok\n",
+ { stderr => 1 },
+ 'no double free when stashes are blessed into each other');
+
+# test if refgen behaves with autoviv magic
{
my @a;
- $a[1] = "ok 53\n";
- print ${\$_} for @a;
+ $a[1] = "good";
+ my $got;
+ for (@a) {
+ $got .= ${\$_};
+ $got .= ';';
+ }
+ is ($got, ";good;");
}
# This test is the reason for postponed destruction in sv_unref
$a = [1,2,3];
$a = $a->[1];
-print "not " unless $a == 2;
-print "ok 54\n";
-
-sub x::DESTROY {print "ok ", 54 + shift->[0], "\n"}
-{ my $a1 = bless [4],"x";
- my $a2 = bless [3],"x";
- { my $a3 = bless [2],"x";
- my $a4 = bless [1],"x";
+is ($a, 2);
+
+# This test used to coredump. The BEGIN block is important as it causes the
+# op that created the constant reference to be freed. Hence the only
+# reference to the constant string "pass" is in $a. The hack that made
+# sure $a = $a->[1] would work didn't work with references to constants.
+
+
+foreach my $lexical ('', 'my $a; ') {
+ my $expect = "pass\n";
+ my $result = runperl (switches => ['-wl'], stderr => 1,
+ prog => $lexical . 'BEGIN {$a = \q{pass}}; $a = $$a; print $a');
+
+ is ($?, 0);
+ is ($result, $expect);
+}
+
+$test = curr_test();
+sub x::DESTROY {print "ok ", $test + shift->[0], "\n"}
+{ my $a1 = bless [3],"x";
+ my $a2 = bless [2],"x";
+ { my $a3 = bless [1],"x";
+ my $a4 = bless [0],"x";
567;
}
}
+curr_test($test+4);
+
+is (runperl (switches=>['-l'],
+ prog=> 'print 1; print qq-*$\*-;print 1;'),
+ "1\n*\n*\n1\n");
+
+# bug #21347
+
+runperl(prog => 'sub UNIVERSAL::AUTOLOAD { qr// } a->p' );
+is ($?, 0, 'UNIVERSAL::AUTOLOAD called when freeing qr//');
+
+runperl(prog => 'sub UNIVERSAL::DESTROY { warn } bless \$a, A', stderr => 1);
+is ($?, 0, 'warn called inside UNIVERSAL::DESTROY');
+
+
+# bug #22719
+
+runperl(prog => 'sub f { my $x = shift; *z = $x; } f({}); f();');
+is ($?, 0, 'coredump on typeglob = (SvRV && !SvROK)');
+# bug #27268: freeing self-referential typeglobs could trigger
+# "Attempt to free unreferenced scalar" warnings
+is (runperl(
+ prog => 'use Symbol;my $x=bless \gensym,q{t}; print;*$$x=$x',
+ stderr => 1
+), '', 'freeing self-referential typeglob');
+
+# using a regex in the destructor for STDOUT segfaulted because the
+# REGEX pad had already been freed (ithreads build only). The
+# object is required to trigger the early freeing of GV refs to to STDOUT
+
+TODO: {
+ local $TODO = "works but output through pipe is mangled" if $^O eq 'VMS';
+ like (runperl(
+ prog => '$x=bless[]; sub IO::Handle::DESTROY{$_=q{bad};s/bad/ok/;print}',
+ stderr => 1
+ ), qr/^(ok)+$/, 'STDOUT destructor');
+}
+
+{
+ no strict 'refs';
+ $name8 = chr 163;
+ $name_utf8 = $name8 . chr 256;
+ chop $name_utf8;
+
+ is ($$name8, undef, 'Nothing before we start');
+ is ($$name_utf8, undef, 'Nothing before we start');
+ $$name8 = "Pound";
+ is ($$name8, "Pound", 'Accessing via 8 bit symref works');
+ is ($$name_utf8, "Pound", 'Accessing via UTF8 symref works');
+}
+
+{
+ no strict 'refs';
+ $name_utf8 = $name = chr 9787;
+ utf8::encode $name_utf8;
+
+ is (length $name, 1, "Name is 1 char");
+ is (length $name_utf8, 3, "UTF8 representation is 3 chars");
+
+ is ($$name, undef, 'Nothing before we start');
+ is ($$name_utf8, undef, 'Nothing before we start');
+ $$name = "Face";
+ is ($$name, "Face", 'Accessing via Unicode symref works');
+ is ($$name_utf8, undef,
+ 'Accessing via the UTF8 byte sequence gives nothing');
+}
+
+{
+ no strict 'refs';
+ $name1 = "\0Chalk";
+ $name2 = "\0Cheese";
+
+ isnt ($name1, $name2, "They differ");
+
+ is ($$name1, undef, 'Nothing before we start (scalars)');
+ is ($$name2, undef, 'Nothing before we start');
+ $$name1 = "Yummy";
+ is ($$name1, "Yummy", 'Accessing via the correct name works');
+ is ($$name2, undef,
+ 'Accessing via a different NUL-containing name gives nothing');
+ # defined uses a different code path
+ ok (defined $$name1, 'defined via the correct name works');
+ ok (!defined $$name2,
+ 'defined via a different NUL-containing name gives nothing');
+
+ is ($name1->[0], undef, 'Nothing before we start (arrays)');
+ is ($name2->[0], undef, 'Nothing before we start');
+ $name1->[0] = "Yummy";
+ is ($name1->[0], "Yummy", 'Accessing via the correct name works');
+ is ($name2->[0], undef,
+ 'Accessing via a different NUL-containing name gives nothing');
+ ok (defined $name1->[0], 'defined via the correct name works');
+ ok (!defined$name2->[0],
+ 'defined via a different NUL-containing name gives nothing');
+
+ my (undef, $one) = @{$name1}[2,3];
+ my (undef, $two) = @{$name2}[2,3];
+ is ($one, undef, 'Nothing before we start (array slices)');
+ is ($two, undef, 'Nothing before we start');
+ @{$name1}[2,3] = ("Very", "Yummy");
+ (undef, $one) = @{$name1}[2,3];
+ (undef, $two) = @{$name2}[2,3];
+ is ($one, "Yummy", 'Accessing via the correct name works');
+ is ($two, undef,
+ 'Accessing via a different NUL-containing name gives nothing');
+ ok (defined $one, 'defined via the correct name works');
+ ok (!defined $two,
+ 'defined via a different NUL-containing name gives nothing');
+
+ is ($name1->{PWOF}, undef, 'Nothing before we start (hashes)');
+ is ($name2->{PWOF}, undef, 'Nothing before we start');
+ $name1->{PWOF} = "Yummy";
+ is ($name1->{PWOF}, "Yummy", 'Accessing via the correct name works');
+ is ($name2->{PWOF}, undef,
+ 'Accessing via a different NUL-containing name gives nothing');
+ ok (defined $name1->{PWOF}, 'defined via the correct name works');
+ ok (!defined $name2->{PWOF},
+ 'defined via a different NUL-containing name gives nothing');
+
+ my (undef, $one) = @{$name1}{'SNIF', 'BEEYOOP'};
+ my (undef, $two) = @{$name2}{'SNIF', 'BEEYOOP'};
+ is ($one, undef, 'Nothing before we start (hash slices)');
+ is ($two, undef, 'Nothing before we start');
+ @{$name1}{'SNIF', 'BEEYOOP'} = ("Very", "Yummy");
+ (undef, $one) = @{$name1}{'SNIF', 'BEEYOOP'};
+ (undef, $two) = @{$name2}{'SNIF', 'BEEYOOP'};
+ is ($one, "Yummy", 'Accessing via the correct name works');
+ is ($two, undef,
+ 'Accessing via a different NUL-containing name gives nothing');
+ ok (defined $one, 'defined via the correct name works');
+ ok (!defined $two,
+ 'defined via a different NUL-containing name gives nothing');
+
+ $name1 = "Left"; $name2 = "Left\0Right";
+ my $glob2 = *{$name2};
+
+ is ($glob1, undef, "We get different typeglobs. In fact, undef");
+
+ *{$name1} = sub {"One"};
+ *{$name2} = sub {"Two"};
+
+ is (&{$name1}, "One");
+ is (&{$name2}, "Two");
+}
+
+# test derefs after list slice
+
+is ( ({foo => "bar"})[0]{foo}, "bar", 'hash deref from list slice w/o ->' );
+is ( ({foo => "bar"})[0]->{foo}, "bar", 'hash deref from list slice w/ ->' );
+is ( ([qw/foo bar/])[0][1], "bar", 'array deref from list slice w/o ->' );
+is ( ([qw/foo bar/])[0]->[1], "bar", 'array deref from list slice w/ ->' );
+is ( (sub {"bar"})[0](), "bar", 'code deref from list slice w/o ->' );
+is ( (sub {"bar"})[0]->(), "bar", 'code deref from list slice w/ ->' );
+
+# deref on empty list shouldn't autovivify
+{
+ local $@;
+ eval { ()[0]{foo} };
+ like ( "$@", "Can't use an undefined value as a HASH reference",
+ "deref of undef from list slice fails" );
+}
+
+# test dereferencing errors
+{
+ format STDERR =
+.
+ my $ref;
+ foreach $ref (*STDOUT{IO}, *STDERR{FORMAT}) {
+ eval q/ $$ref /;
+ like($@, qr/Not a SCALAR reference/, "Scalar dereference");
+ eval q/ @$ref /;
+ like($@, qr/Not an ARRAY reference/, "Array dereference");
+ eval q/ %$ref /;
+ like($@, qr/Not a HASH reference/, "Hash dereference");
+ eval q/ &$ref /;
+ like($@, qr/Not a CODE reference/, "Code dereference");
+ }
+
+ $ref = *STDERR{FORMAT};
+ eval q/ *$ref /;
+ like($@, qr/Not a GLOB reference/, "Glob dereference");
+
+ $ref = *STDOUT{IO};
+ eval q/ *$ref /;
+ is($@, '', "Glob dereference of PVIO is acceptable");
+
+ is($ref, *{$ref}{IO}, "IO slot of the temporary glob is set correctly");
+}
+
+# these will segfault if they fail
+
+my $pvbm = PVBM;
+my $rpvbm = \$pvbm;
+
+ok (!eval { *$rpvbm }, 'PVBM ref is not a GLOB ref');
+ok (!eval { *$pvbm }, 'PVBM is not a GLOB ref');
+ok (!eval { $$pvbm }, 'PVBM is not a SCALAR ref');
+ok (!eval { @$pvbm }, 'PVBM is not an ARRAY ref');
+ok (!eval { %$pvbm }, 'PVBM is not a HASH ref');
+ok (!eval { $pvbm->() }, 'PVBM is not a CODE ref');
+ok (!eval { $rpvbm->foo }, 'PVBM is not an object');
+
+# bug 24254
+is( runperl(stderr => 1, prog => 'map eval qq(exit),1 for 1'), "");
+is( runperl(stderr => 1, prog => 'eval { for (1) { map { die } 2 } };'), "");
+is( runperl(stderr => 1, prog => 'for (125) { map { exit } (213)}'), "");
+my $hushed = $^O eq 'VMS' ? 'use vmsish qw(hushed);' : '';
+is( runperl(stderr => 1, prog => $hushed . 'map die,4 for 3'), "Died at -e line 1.\n");
+is( runperl(stderr => 1, prog => $hushed . 'grep die,4 for 3'), "Died at -e line 1.\n");
+is( runperl(stderr => 1, prog => $hushed . 'for $a (3) {@b=sort {die} 4,5}'), "Died at -e line 1.\n");
+
+# bug 57564
+is( runperl(stderr => 1, prog => 'my $i;for $i (1) { for $i (2) { } }'), "");
+
+# The mechanism for freeing objects in globs used to leave dangling
+# pointers to freed SVs. To test this, we construct this nested structure:
+# GV => blessed(AV) => RV => GV => blessed(SV)
+# all with a refcnt of 1, and hope that the second GV gets processed first
+# by do_clean_named_objs. Then when the first GV is processed, it mustn't
+# find anything nasty left by the previous GV processing.
+# The eval is stop things in the main body of the code holding a reference
+# to a GV, and the print at the end seems to bee necessary to ensure
+# the correct freeing order of *x and *y (no, I don't know why - DAPM).
+
+is (runperl(
+ prog => 'eval q[bless \@y; bless \$x; $y[0] = \*x; $z = \*y; ]; '
+ . 'delete $::{x}; delete $::{y}; print qq{ok\n};',
+ stderr => 1),
+ "ok\n", 'freeing freed glob in global destruction');
+
+
+# Test undefined hash references as arguments to %{} in boolean context
+# [perl #81750]
+{
+ no strict 'refs';
+ eval { my $foo; %$foo; }; ok !$@, '%$undef';
+ eval { my $foo; scalar %$foo; }; ok !$@, 'scalar %$undef';
+ eval { my $foo; !%$foo; }; ok !$@, '!%$undef';
+ eval { my $foo; if ( %$foo) {} }; ok !$@, 'if ( %$undef) {}';
+ eval { my $foo; if (!%$foo) {} }; ok !$@, 'if (!%$undef) {}';
+ eval { my $foo; unless ( %$foo) {} }; ok !$@, 'unless ( %$undef) {}';
+ eval { my $foo; unless (!%$foo) {} }; ok !$@, 'unless (!%$undef) {}';
+ eval { my $foo; 1 if %$foo; }; ok !$@, '1 if %$undef';
+ eval { my $foo; 1 if !%$foo; }; ok !$@, '1 if !%$undef';
+ eval { my $foo; 1 unless %$foo; }; ok !$@, '1 unless %$undef;';
+ eval { my $foo; 1 unless ! %$foo; }; ok !$@, '1 unless ! %$undef';
+ eval { my $foo; %$foo ? 1 : 0; }; ok !$@, ' %$undef ? 1 : 0';
+ eval { my $foo; !%$foo ? 1 : 0; }; ok !$@, '!%$undef ? 1 : 0';
+}
+
+# RT #88330
+# Make sure that a leaked thinggy with multiple weak references to
+# it doesn't trigger a panic with multiple rounds of global cleanup
+# (Perl_sv_clean_all).
+
+SKIP: {
+ skip_if_miniperl('no Scalar::Util under miniperl', 4);
+
+ local $ENV{PERL_DESTRUCT_LEVEL} = 2;
+
+ # we do all permutations of array/hash, 1ref/2ref, to account
+ # for the different way backref magic is stored
+
+ fresh_perl_is(<<'EOF', 'ok', { stderr => 1 }, 'array with 1 weak ref');
+use Scalar::Util qw(weaken);
+my $r = [];
+Internals::SvREFCNT(@$r, 9);
+my $r1 = $r;
+weaken($r1);
+print "ok";
+EOF
+
+ fresh_perl_is(<<'EOF', 'ok', { stderr => 1 }, 'array with 2 weak refs');
+use Scalar::Util qw(weaken);
+my $r = [];
+Internals::SvREFCNT(@$r, 9);
+my $r1 = $r;
+weaken($r1);
+my $r2 = $r;
+weaken($r2);
+print "ok";
+EOF
+
+ fresh_perl_is(<<'EOF', 'ok', { stderr => 1 }, 'hash with 1 weak ref');
+use Scalar::Util qw(weaken);
+my $r = {};
+Internals::SvREFCNT(%$r, 9);
+my $r1 = $r;
+weaken($r1);
+print "ok";
+EOF
+
+ fresh_perl_is(<<'EOF', 'ok', { stderr => 1 }, 'hash with 2 weak refs');
+use Scalar::Util qw(weaken);
+my $r = {};
+Internals::SvREFCNT(%$r, 9);
+my $r1 = $r;
+weaken($r1);
+my $r2 = $r;
+weaken($r2);
+print "ok";
+EOF
+
+}
+
+SKIP:{
+ skip_if_miniperl "no Scalar::Util on miniperl", 1;
+ my $error;
+ *hassgropper::DESTROY = sub {
+ require Scalar::Util;
+ eval { Scalar::Util::weaken($_[0]) };
+ $error = $@;
+ # This line caused a crash before weaken refused to weaken a
+ # read-only reference:
+ $do::not::overwrite::this = $_[0];
+ };
+ my $xs = bless [], "hassgropper";
+ undef $xs;
+ like $error, qr/^Modification of a read-only/,
+ 'weaken refuses to weaken a read-only ref';
+ # Now that the test has passed, avoid sabotaging global destruction:
+ undef *hassgropper::DESTROY;
+ undef $do::not::overwrite::this;
+}
+
+
+is ref( bless {}, "nul\0clean" ), "nul\0clean", "ref() is nul-clean";
+
+# Test constants and references thereto.
+for (3) {
+ eval { $_ = 4 };
+ like $@, qr/^Modification of a read-only/,
+ 'assignment to value aliased to literal number';
+ require Config;
+ local $::TODO = " " if $Config::Config{useithreads};
+ eval { ${\$_} = 4 };
+ like $@, qr/^Modification of a read-only/,
+ 'refgen does not allow assignment to value aliased to literal number';
+}
+for ("4eounthouonth") {
+ eval { $_ = 4 };
+ like $@, qr/^Modification of a read-only/,
+ 'assignment to value aliased to literal string';
+ require Config;
+ local $::TODO = " " if $Config::Config{useithreads};
+ eval { ${\$_} = 4 };
+ like $@, qr/^Modification of a read-only/,
+ 'refgen does not allow assignment to value aliased to literal string';
+}
+{
+ local $::TODO = ' '
+ if $Config::Config{useithreads} && $Config::Config{mad};
+ my $aref = \123;
+ is \$$aref, $aref,
+ '[perl #109746] referential identity of \literal under threads+mad'
+}
+
+# Bit of a hack to make test.pl happy. There are 3 more tests after it leaves.
+$test = curr_test();
+curr_test($test + 3);
# test global destruction
-my $test = 59;
my $test1 = $test + 1;
my $test2 = $test + 2;
DESTROY {
print $_[0][0];
}
+
https://perl5.git.perl.org / perl5.git / blobdiff