chdir 't' if -d 't';
@INC = '../lib';
}
+our $Message = "Noname test";
eval 'use Config'; # Defaults assumed if this fails
+run_tests() unless caller;
+
+sub run_tests {
+
$x = "abc\ndef\n";
if ($x =~ /^abc/) {print "ok 1\n";} else {print "not ok 1\n";}
print "ok $test\n";
$test++;
+if ($::running_as_thread) {
+ print "not ok $test # TODO & SKIP: croaks in 5.10 when threaded\n";
+ $test++;
+} else {
$a=qr/(?{++$b})/;
$b = 7;
/$a$a/;
print "not " unless $b eq '9';
print "ok $test\n";
$test++;
+}
-$c="$a";
-/$a$a/;
-print "not " unless $b eq '11';
-print "ok $test\n";
-$test++;
+{
+ local $TODO = $::running_as_thread;
+ $c="$a";
+ /$a$a/;
+ iseq($b, '11');
+}
{
use re "eval";
/$a$c$a/;
- print "not " unless $b eq '14';
- print "ok $test\n";
- $test++;
+ {
+ local $TODO = $::running_as_thread;
+ iseq($b, '14');
+ }
local $lex_a = 2;
my $lex_a = 43;
no re "eval";
$match = eval { /$a$c$a/ };
- print "not "
- unless $b eq '14' and $@ =~ /Eval-group not allowed/ and not $match;
- print "ok $test\n";
- $test++;
+ # FIXME - split this one. That would require removing a lot of hard coded
+ # test numbers.
+ local $TODO = $::running_as_thread;
+ ok($b eq '14' and $@ =~ /Eval-group not allowed/ and not $match);
}
{
print "ok $test\n";
$test++;
-print "not " unless $str =~ /.\G./ and $& eq 'bc';
-print "ok $test\n";
-$test++;
+{
+ local $TODO = $::running_as_thread;
+ ok($str =~ /.\G./ and $& eq 'bc');
+}
print "not " unless $str =~ /\G../ and $& eq 'cd';
print "ok $test\n";
pos($foo)=1;
$foo=~/.\G(..)/g;
-print "not " unless($1 eq 'ab');
-print "ok $test\n";
-$test++;
+{
+ local $TODO = $::running_as_thread;
+ iseq($1,'ab');
+}
pos($foo) += 1;
$foo=~/.\G(..)/g;
-print "not " unless($1 eq 'cc');
-print "ok $test\n";
-$test++;
+{
+ local $TODO = $::running_as_thread;
+ iseq($1, 'cc');
+}
pos($foo) += 1;
$foo=~/.\G(..)/g;
-print "not " unless($1 eq 'de');
-print "ok $test\n";
-$test++;
+{
+ local $TODO = $::running_as_thread;
+ iseq($1, 'de');
+}
-print "not " unless $foo =~ /\Gef/g;
-print "ok $test\n";
-$test++;
+{
+ local $TODO = $::running_as_thread;
+ ok($foo =~ /\Gef/g);
+}
undef pos $foo;
@b = grep(/\w/,@a);
@c = grep(/[\w]/,@a);
-print "not " if "@b" ne "@c";
-print "ok $test\n";
-$test++;
+iseq("@b","@c");
# see if backtracking optimization works correctly
"\n\n" =~ /\n $ \n/x or print "not ";
# Force scalar context on the patern match
sub ok ($;$) {
my($ok, $name) = @_;
+ my $todo = $TODO ? " # TODO $TODO" : '';
- printf "%sok %d - %s\n", ($ok ? "" : "not "), $test, $name||'unnamed';
+ printf "%sok %d - %s\n", ($ok ? "" : "not "), $test,
+ ($name||$Message)."$todo\tLine ".((caller)[2]);
printf "# Failed test at line %d\n", (caller)[2] unless $ok;
ok($a !~ /^\C{4}y/, q{don't match \C{4}y});
}
-$_ = 'aaaaaaaaaa';
-utf8::upgrade($_); chop $_; $\="\n";
-ok(/[^\s]+/, "m/[^\s]/ utf8");
-ok(/[^\d]+/, "m/[^\d]/ utf8");
-ok(($a = $_, $_ =~ s/[^\s]+/./g), "s/[^\s]/ utf8");
-ok(($a = $_, $a =~ s/[^\d]+/./g), "s/[^\s]/ utf8");
+{
+ local $\;
+ $_ = 'aaaaaaaaaa';
+ utf8::upgrade($_); chop $_; $\="\n";
+ ok(/[^\s]+/, "m/[^\s]/ utf8");
+ ok(/[^\d]+/, "m/[^\d]/ utf8");
+ ok(($a = $_, $_ =~ s/[^\s]+/./g), "s/[^\s]/ utf8");
+ ok(($a = $_, $a =~ s/[^\d]+/./g), "s/[^\s]/ utf8");
+}
ok("\x{100}" =~ /\x{100}/, "[perl #15397]");
ok("\x{100}" =~ /(\x{100})/, "[perl #15397]");
foreach (1,2,3,4) {
$p++ if /(??{ $p })/
}
- ok ($p == 5, "[perl #20683] (??{ }) returns stale values");
+ iseq ($p, 5, "[perl #20683] (??{ }) returns stale values");
{ package P; $a=1; sub TIESCALAR { bless[] } sub FETCH { $a++ } }
tie $p, P;
foreach (1,2,3,4) {
/(??{ $p })/
}
- ok ( $p == 5, "(??{ }) returns stale values");
+ iseq ( $p, 5, "(??{ }) returns stale values");
}
{
#$_ = "x"; /x(?{func "in regexp"})/;
#$_ = "x"; /x(?{func "in multiline regexp"})/m;
-# bug #19049
+# bug RT#19049
$_="abcdef\n";
@x = m/./g;
-ok("abcde" eq "$`", '# TODO #19049 - global match not setting $`');
+ok("abcde" eq "$`", 'RT#19049 - global match not setting $`');
ok("123\x{100}" =~ /^.*1.*23\x{100}$/, 'uft8 + multiple floating substr');
}
-
+print "# set PERL_SKIP_PSYCHO_TEST to skip this test\n";
if (!$ENV{PERL_SKIP_PSYCHO_TEST}){
my @normal=qw(these are some normal words);
my $psycho=join "|",@normal,map chr $_,255..20000;
ok($utf8 =~ /(abc|\xe9)/i, "utf8/latin trie");
ok($utf8 =~ /(abc|$latin1)/i, "utf8/latin trie runtime");
- ok("\xe9" =~ /$utf8/i, "# TODO latin/utf8");
+ ok("\xe9" =~ /$utf8/i, "# latin/utf8");
ok("\xe9" =~ /(abc|$utf8)/i, "# latin/utf8 trie");
- ok($latin1 =~ /$utf8/i, "# TODO latin/utf8 runtime");
+ ok($latin1 =~ /$utf8/i, "# latin/utf8 runtime");
ok($latin1 =~ /(abc|$utf8)/i, "# latin/utf8 trie runtime");
}
}
{
+ local $TODO = "See changes 26925-26928, which reverted change 26410";
package lv;
$var = "abc";
sub variable : lvalue { $var }
my $o = bless [], "lv";
my $f = "";
eval { for (1..2) { $f .= $1 if $o->variable =~ /(.)/g } };
- ok($f eq "ab", "pos retained between calls # TODO") or print "# $@\n";
+ ok($f eq "ab", "pos retained between calls") or print "# $@\n";
}
{
+ local $TODO = "See changes 26925-26928, which reverted change 26410";
$var = "abc";
sub variable : lvalue { $var }
my $f = "";
eval { for (1..2) { $f .= $1 if variable() =~ /(.)/g } };
- ok($f eq "ab", "pos retained between calls # TODO") or print "# $@\n";
+ ok($f eq "ab", "pos retained between calls") or print "# $@\n";
}
# [perl #37836] Simple Regex causes SEGV when run on specific data
$s=~s/(?'digits'\d+)\k'digits'/$+{digits}/;
ok($s eq '123456','Named capture (single quotes) s///');
}
+
+{
+ my @ary = (
+ pack('U', 0x00F1), # n-tilde
+ '_'.pack('U', 0x00F1), # _ + n-tilde
+ 'c'.pack('U', 0x0327), # c + cedilla
+ pack('U*', 0x00F1, 0x0327), # n-tilde + cedilla
+ 'a'.pack('U', 0x00B2), # a + superscript two
+ pack('U', 0x0391), # ALPHA
+ pack('U', 0x0391).'2', # ALPHA + 2
+ pack('U', 0x0391).'_', # ALPHA + _
+ );
+ for my $uni (@ary) {
+ my ($r1, $c1, $r2, $c2) = eval qq{
+ use utf8;
+ scalar("..foo foo.." =~ /(?'${uni}'foo) \\k'${uni}'/),
+ \$+{${uni}},
+ scalar("..bar bar.." =~ /(?<${uni}>bar) \\k<${uni}>/),
+ \$+{${uni}};
+ };
+ ok($r1, "Named capture UTF (?'')");
+ ok(defined $c1 && $c1 eq 'foo', "Named capture UTF \%+");
+ ok($r2, "Named capture UTF (?<>)");
+ ok(defined $c2 && $c2 eq 'bar', "Named capture UTF \%+");
+ }
+}
+
sub iseq($$;$) {
my ( $got, $expect, $name)=@_;
+ my $todo = $TODO ? " # TODO $TODO" : '';
$_=defined($_) ? "'$_'" : "undef"
for $got, $expect;
my $ok= $got eq $expect;
- printf "%sok %d - %s\n", ($ok ? "" : "not "), $test, $name||'unnamed';
+ printf "%sok %d - %s$todo\n", ($ok ? "" : "not "), $test,
+ ($name||$Message)."\tLine ".((caller)[2]);
printf "# Failed test at line %d\n".
"# expected: %s\n".
';
ok(!$@,'lvalue $+{...} should not throw an exception');
}
-
+{
+ my $s='foo bar baz';
+ my @res;
+ if ('1234'=~/(?<A>1)(?<B>2)(?<A>3)(?<B>4)/) {
+ foreach my $name (sort keys(%-)) {
+ my $ary = $-{$name};
+ foreach my $idx (0..$#$ary) {
+ push @res,"$name:$idx:$ary->[$idx]";
+ }
+ }
+ }
+ my @expect=qw(A:0:1 A:1:3 B:0:2 B:1:4);
+ iseq("@res","@expect","Check %-");
+ eval'
+ print for $-{this_key_doesnt_exist};
+ ';
+ ok(!$@,'lvalue $-{...} should not throw an exception');
+}
# stress test CURLYX/WHILEM.
#
# This test includes varying levels of nesting, and according to
if ($ENV{PERL_SKIP_PSYCHO_TEST}){
printf "ok %d Skip: No psycho tests\n", $test++;
} else {
+ print "# set PERL_SKIP_PSYCHO_TEST to skip this test\n";
my $r = qr/^
(?:
( (?:a|z+)+ )
}
{
# Test named commits and the $REGERROR var
+ local $Message = "\$REGERROR";
our $REGERROR;
for $word (qw(bar baz bop)) {
$REGERROR="";
}
}
{ #Regression test for perlbug 40684
+ local $Message = "RT#40684 tests:";
my $s = "abc\ndef";
my $rex = qr'^abc$'m;
ok($s =~ m/$rex/);
}
{
+ local $Message = "Relative Recursion";
my $parens=qr/(\((?:[^()]++|(?-1))*+\))/;
local $_='foo((2*3)+4-3) + bar(2*(3+4)-1*(2-3))';
my ($all,$one,$two)=('','','');
iseq($all, 'foo((2*3)+4-3) + bar(2*(3+4)-1*(2-3))');
iseq($all, $_);
}
+{
+ my $spaces=" ";
+ local $_=join 'bar',$spaces,$spaces;
+ our $count=0;
+ s/(?>\s+bar)(?{$count++})//g;
+ iseq($_,$spaces,"SUSPEND final string");
+ iseq($count,1,"Optimiser should have prevented more than one match");
+}
+{
+ local $Message = "RT#36909 test";
+ $^R = 'Nothing';
+ {
+ local $^R = "Bad";
+ ok('x foofoo y' =~ m{
+ (foo) # $^R correctly set
+ (?{ "last regexp code result" })
+ }x);
+ iseq($^R,'last regexp code result');
+ }
+ iseq($^R,'Nothing');
+ {
+ local $^R = "Bad";
-#-------------------------------------------------------------------
+ ok('x foofoo y' =~ m{
+ (?:foo|bar)+ # $^R correctly set
+ (?{"last regexp code result"})
+ }x);
+ iseq($^R,'last regexp code result');
+ }
+ iseq($^R,'Nothing');
+
+ {
+ local $^R = "Bad";
+ ok('x foofoo y' =~ m{
+ (foo|bar)\1+ # $^R undefined
+ (?{"last regexp code result"})
+ }x);
+ iseq($^R,'last regexp code result');
+ }
+ iseq($^R,'Nothing');
+
+ {
+ local $^R = "Bad";
+ ok('x foofoo y' =~ m{
+ (foo|bar)\1 # this time without the +
+ (?{"last regexp code result"})
+ }x);
+ iseq($^R,'last regexp code result');
+ }
+ iseq($^R,'Nothing');
+}
+{
+ local $Message="RT 22395";
+ local $TODO = "Should be L+1 not L*(L+3)/2 (L=$l)";
+ our $count;
+ for my $l (10,100,1000) {
+ $count=0;
+ ('a' x $l) =~ /(.*)(?{$count++})[bc]/;
+ iseq( $count, $l + 1);
+ }
+}
+{
+ local $Message = "RT#22614";
+ local $_='ab';
+ our @len=();
+ /(.){1,}(?{push @len,0+@-})(.){1,}(?{})^/;
+ iseq("@len","2 2 2");
+}
+{
+ local $Message = "RT#18209";
+ my $text = ' word1 word2 word3 word4 word5 word6 ';
+
+ my @words = ('word1', 'word3', 'word5');
+ my $count;
+ foreach my $word (@words){
+ $text =~ s/$word\s//gi; # Leave a space to seperate words in the resultant str.
+ # The following block is not working.
+ if($&){
+ $count++;
+ }
+ # End bad block
+ }
+ iseq($count,3);
+ iseq($text,' word2 word4 word6 ');
+}
+{
+ # RT#6893
+ local $_= qq(A\nB\nC\n);
+ my @res;
+ while (m#(\G|\n)([^\n]*)\n#gsx)
+ {
+ push @res,"$2";
+ last if @res>3;
+ }
+ iseq("@res","A B C","RT#6893: /g pattern shouldn't infinite loop");
+}
+
+{
+ # From Message-ID: <877ixs6oa6.fsf@k75.linux.bogus>
+ my $dow_name= "nada";
+ my $parser = "(\$dow_name) = \$time_string =~ /(D\x{e9}\\ C\x{e9}adaoin|D\x{e9}\\ Sathairn|\\w+|\x{100})/";
+ my $time_string = "D\x{e9} C\x{e9}adaoin";
+ eval $parser;
+ ok(!$@,"Test Eval worked");
+ iseq($dow_name,$time_string,"UTF8 trie common prefix extraction");
+}
+
+{
+ my $v;
+ ($v='bar')=~/(\w+)/g;
+ $v='foo';
+ iseq("$1",'bar','$1 is safe after /g - may fail due to specialized config in pp_hot.c')
+}
+{
+ local $Message = "http://nntp.perl.org/group/perl.perl5.porters/118663";
+ my $qr_barR1 = qr/(bar)\g-1/;
+ ok("foobarbarxyz" =~ $qr_barR1);
+ ok("foobarbarxyz" =~ qr/foo${qr_barR1}xyz/);
+ ok("foobarbarxyz" =~ qr/(foo)${qr_barR1}xyz/);
+ ok("foobarbarxyz" =~ qr/(foo)(bar)\g{-1}xyz/);
+ ok("foobarbarxyz" =~ qr/(foo${qr_barR1})xyz/);
+ ok("foobarbarxyz" =~ qr/(foo(bar)\g{-1})xyz/);
+}
+{
+ local $Message = "RT#41010";
+ my @tails=('','(?(1))','(|)','()?');
+ my @quants=('*','+');
+ my $doit=sub {
+ my $pats= shift;
+ for (@_) {
+ for my $pat (@$pats) {
+ for my $quant (@quants) {
+ for my $tail (@tails) {
+ my $re = "($pat$quant\$)$tail";
+ ok(/$re/ && $1 eq $_,"'$_'=~/$re/");
+ ok(/$re/m && $1 eq $_,"'$_'=~/$re/m");
+ }
+ }
+ }
+ }
+ };
+
+ my @dpats=(
+ '\d',
+ '[1234567890]',
+ '(1|[23]|4|[56]|[78]|[90])',
+ '(?:1|[23]|4|[56]|[78]|[90])',
+ '(1|2|3|4|5|6|7|8|9|0)',
+ '(?:1|2|3|4|5|6|7|8|9|0)',
+ );
+ my @spats=('[ ]',' ','( |\t)','(?: |\t)','[ \t]','\s');
+ my @sstrs=(' ');
+ my @dstrs=('12345');
+ $doit->(\@spats,@sstrs);
+ $doit->(\@dpats,@dstrs);
+}
+{
+ local $Message = "\$REGMARK";
+ our @r=();
+ ok('foofoo' =~ /foo (*MARK:foo) (?{push @r,$REGMARK}) /x);
+ iseq("@r","foo");
+ iseq($REGMARK,"foo");
+ ok('foofoo' !~ /foo (*MARK:foo) (*FAIL) /x);
+ ok(!$REGMARK);
+ iseq($REGERROR,'foo');
+}
+{
+ my $x;
+ $x = "abc.def.ghi.jkl";
+ $x =~ s/.*\K\..*//;
+ ok($x eq "abc.def.ghi");
+
+ $x = "one two three four";
+ $x =~ s/o+ \Kthree//g;
+ ok($x eq "one two four");
+
+ $x = "abcde";
+ $x =~ s/(.)\K/$1/g;
+ ok($x eq "aabbccddee");
+}
+sub kt
+{
+ return '4' if $_[0] eq '09028623';
+}
+
+{ # Nested EVAL using PL_curpm (via $1 or friends)
+ my $re;
+ our $grabit = qr/ ([0-6][0-9]{7}) (??{ kt $1 }) [890] /x;
+ $re = qr/^ ( (??{ $grabit }) ) $ /x;
+ my @res = '0902862349' =~ $re;
+ iseq(join("-",@res),"0902862349",
+ 'PL_curpm is set properly on nested eval');
+
+ our $qr = qr/ (o) (??{ $1 }) /x;
+ ok( 'boob'=~/( b (??{ $qr }) b )/x && 1,
+ "PL_curpm, nested eval");
+}
+
+{
+ use charnames ":full";
+ ok("\N{ROMAN NUMERAL ONE}" =~ /\p{Alphabetic}/, "I =~ Alphabetic");
+ ok("\N{ROMAN NUMERAL ONE}" =~ /\p{Uppercase}/, "I =~ Uppercase");
+ ok("\N{ROMAN NUMERAL ONE}" !~ /\p{Lowercase}/, "I !~ Lowercase");
+ ok("\N{ROMAN NUMERAL ONE}" =~ /\p{IDStart}/, "I =~ ID_Start");
+ ok("\N{ROMAN NUMERAL ONE}" =~ /\p{IDContinue}/, "I =~ ID_Continue");
+ ok("\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{Alphabetic}/, "i =~ Alphabetic");
+ ok("\N{SMALL ROMAN NUMERAL ONE}" !~ /\p{Uppercase}/, "i !~ Uppercase");
+ ok("\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{Lowercase}/, "i =~ Lowercase");
+ ok("\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{IDStart}/, "i =~ ID_Start");
+ ok("\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{IDContinue}/, "i =~ ID_Continue");
+}
+
+{
+# requirement of Unicode Technical Standard #18, 1.7 Code Points
+# cf. http://www.unicode.org/reports/tr18/#Supplementary_Characters
+ for my $u (0x7FF, 0x800, 0xFFFF, 0x10000) {
+ no warnings 'utf8'; # oops
+ my $c = chr $u;
+ my $x = sprintf '%04X', $u;
+ ok( "A${c}B" =~ /A[\0-\x{10000}]B/, "unicode range - $x");
+ }
+}
+
+{
+ my $res="";
+
+ if ('1' =~ /(?|(?<digit>1)|(?<digit>2))/) {
+ $res = "@{$- {digit}}";
+ }
+ iseq($res,"1",
+ "Check that (?|...) doesnt cause dupe entries in the names array");
+ #---
+ $res="";
+ if ('11' =~ /(?|(?<digit>1)|(?<digit>2))(?&digit)/) {
+ $res = "@{$- {digit}}";
+ }
+ iseq($res, "1",
+ "Check that (?&..) to a buffer inside a (?|...) goes to the leftmost");
+}
+{
+ use warnings;
+ local $Message = "ASCII pattern that really is utf8";
+ my @w;
+ local $SIG{__WARN__}=sub{push @w,"@_"};
+ my $c=qq(\x{DF});
+ ok($c=~/${c}|\x{100}/);
+ ok(@w==0);
+}
+{
+ local $Message = "corruption of match results of qr// across scopes";
+ my $qr=qr/(fo+)(ba+r)/;
+ 'foobar'=~/$qr/;
+ iseq("$1$2","foobar");
+ {
+ 'foooooobaaaaar'=~/$qr/;
+ iseq("$1$2",'foooooobaaaaar');
+ }
+ iseq("$1$2","foobar");
+}
+{
+ local $Message = "HORIZWS";
+ local $_="\t \r\n \n \t".chr(11)."\n";
+ s/\H/H/g;
+ s/\h/h/g;
+ iseq($_,"hhHHhHhhHH");
+ $_="\t \r\n \n \t".chr(11)."\n";
+ utf8::upgrade($_);
+ s/\H/H/g;
+ s/\h/h/g;
+ iseq($_,"hhHHhHhhHH");
+}
+{
+ local $Message = "Various whitespace special patterns";
+ my @h=map { chr( $_ ) } (
+ 0x09, 0x20, 0xa0, 0x1680, 0x180e, 0x2000, 0x2001, 0x2002,
+ 0x2003, 0x2004, 0x2005, 0x2006, 0x2007, 0x2008, 0x2009, 0x200a,
+ 0x202f, 0x205f, 0x3000
+ );
+ my @v=map { chr( $_ ) } ( 0x0a, 0x0b, 0x0c, 0x0d, 0x85, 0x2028, 0x2029 );
+ my @lb=( "\x0D\x0A",
+ map { chr( $_ ) } ( 0x0A..0x0D,0x85,0x2028,0x2029 ));
+ foreach my $t ([\@h,qr/\h/,qr/\h+/],[\@v,qr/\v/,qr/\v+/],[\@lb,qr/\R/,qr/\R+/],){
+ my $ary=shift @$t;
+ foreach my $pat (@$t) {
+ foreach my $str (@$ary) {
+ ok($str=~/($pat)/,$pat);
+ iseq($1,$str,$pat);
+ utf8::upgrade($str);
+ ok($str=~/($pat)/,"Upgraded string - $pat");
+ iseq($1,$str,"Upgraded string - $pat");
+ }
+ }
+ }
+}
+{
+ local $Message = "Check that \\xDF match properly in its various forms";
+ # test that \xDF matches properly. this is pretty hacky stuff,
+ # but its actually needed. the malarky with '-' is to prevent
+ # compilation caching from playing any role in the test.
+ my @df= (chr(0xDF),'-',chr(0xDF));
+ utf8::upgrade($df[2]);
+ my @strs= ('ss','sS','Ss','SS',chr(0xDF));
+ my @ss= map { ("$_", "$_") } @strs;
+ utf8::upgrade($ss[$_*2+1]) for 0..$#strs;
+
+ for my $ssi (0..$#ss) {
+ for my $dfi (0..$#df) {
+ my $pat= $df[$dfi];
+ my $str= $ss[$ssi];
+ my $utf_df= ($dfi > 1) ? 'utf8' : '';
+ my $utf_ss= ($ssi % 2) ? 'utf8' : '';
+ (my $sstr=$str)=~s/\xDF/\\xDF/;
+
+ if ($utf_df || $utf_ss || length($ss[$ssi])==1) {
+ my $ret= $str=~/$pat/i;
+ next if $pat eq '-';
+ ok($ret,
+ "\"$sstr\"=~/\\xDF/i (str is @{[$utf_ss||'latin']}, pat is @{[$utf_df||'latin']})");
+ } else {
+ my $ret= $str !~ /$pat/i;
+ next if $pat eq '-';
+ ok($ret,
+ "\"$sstr\"!~/\\xDF/i (str is @{[$utf_ss||'latin']}, pat is @{[$utf_df||'latin']})");
+ }
+ }
+ }
+}
+{
+ local $Message = "BBC(Bleadperl Breaks CPAN) Today: String::Multibyte";
+ my $re = qr/(?:[\x00-\xFF]{4})/;
+ my $hyp = "\0\0\0-";
+ my $esc = "\0\0\0\\";
+
+ my $str = "$esc$hyp$hyp$esc$esc";
+ my @a = ($str =~ /\G(?:\Q$esc$esc\E|\Q$esc$hyp\E|$re)/g);
+
+ iseq(0+@a,3);
+ iseq(join('=', @a),"$esc$hyp=$hyp=$esc$esc");
+}
+# test for keys in %+ and %-
+{
+ my $_ = "abcdef";
+ /(?<foo>a)|(?<foo>b)/;
+ iseq( (join ",", sort keys %+), "foo" );
+ iseq( (join ",", sort keys %-), "foo" );
+ iseq( (join ",", sort values %+), "a" );
+ iseq( (join ",", sort map "@$_", values %-), "a " );
+ /(?<bar>a)(?<bar>b)(?<quux>.)/;
+ iseq( (join ",", sort keys %+), "bar,quux" );
+ iseq( (join ",", sort keys %-), "bar,quux" );
+ iseq( (join ",", sort values %+), "a,c" ); # leftmost
+ iseq( (join ",", sort map "@$_", values %-), "a b,c" );
+ /(?<un>a)(?<deux>c)?/; # second buffer won't capture
+ iseq( (join ",", sort keys %+), "un" );
+ iseq( (join ",", sort keys %-), "deux,un" );
+ iseq( (join ",", sort values %+), "a" );
+ iseq( (join ",", sort map "@$_", values %-), ",a" );
+}
+
+# length() on captures, the numbered ones end up in Perl_magic_len
+{
+ my $_ = "aoeu \xe6var ook";
+ /^ \w+ \s (?<eek>\S+)/x;
+
+ iseq( length($`), 0, 'length $`' );
+ iseq( length($'), 4, q[length $'] );
+ iseq( length($&), 9, 'length $&' );
+ iseq( length($1), 4, 'length $1' );
+ iseq( length($+{eek}), 4, 'length $+{eek} == length $1' );
+}
+
+{
+ my $ok=-1;
+
+ $ok=exists($-{x}) ? 1 : 0
+ if 'bar'=~/(?<x>foo)|bar/;
+ iseq($ok,1,'$-{x} exists after "bar"=~/(?<x>foo)|bar/');
+ iseq(scalar(%+), 0, 'scalar %+ == 0 after "bar"=~/(?<x>foo)|bar/');
+ iseq(scalar(%-), 1, 'scalar %- == 1 after "bar"=~/(?<x>foo)|bar/');
+
+ $ok=-1;
+ $ok=exists($+{x}) ? 1 : 0
+ if 'bar'=~/(?<x>foo)|bar/;
+ iseq($ok,0,'$+{x} not exists after "bar"=~/(?<x>foo)|bar/');
+ iseq(scalar(%+), 0, 'scalar %+ == 0 after "bar"=~/(?<x>foo)|bar/');
+ iseq(scalar(%-), 1, 'scalar %- == 1 after "bar"=~/(?<x>foo)|bar/');
+
+ $ok=-1;
+ $ok=exists($-{x}) ? 1 : 0
+ if 'foo'=~/(?<x>foo)|bar/;
+ iseq($ok,1,'$-{x} exists after "foo"=~/(?<x>foo)|bar/');
+ iseq(scalar(%+), 1, 'scalar %+ == 1 after "foo"=~/(?<x>foo)|bar/');
+ iseq(scalar(%-), 1, 'scalar %- == 1 after "foo"=~/(?<x>foo)|bar/');
+
+ $ok=-1;
+ $ok=exists($+{x}) ? 1 : 0
+ if 'foo'=~/(?<x>foo)|bar/;
+ iseq($ok,1,'$+{x} exists after "foo"=~/(?<x>foo)|bar/');
+}
+{
+ local $_;
+ ($_ = 'abc')=~/(abc)/g;
+ $_ = '123';
+ iseq("$1",'abc',"/g leads to unsafe match vars: $1");
+}
+{
+ local $Message="Message-ID: <20070818091501.7eff4831@r2d2>";
+ my $str= "";
+ for(0..5){
+ my @x;
+ $str .= "@x"; # this should ALWAYS be the empty string
+ 'a'=~/(a|)/;
+ push @x,1;
+ }
+ iseq(length($str),"0","Trie scope error, string should be empty");
+ $str="";
+ my @foo = ('a')x5;
+ for (@foo) {
+ my @bar;
+ $str .= "@bar";
+ s/a|/push @bar, 1/e;
+ }
+ iseq(length($str),"0","Trie scope error, string should be empty");
+}
+{
+# [perl #45605] Regexp failure with utf8-flagged and byte-flagged string
+
+ my $utf_8 = "\xd6schel";
+ utf8::upgrade($utf_8);
+ $utf_8 =~ m{(\xd6|Ö)schel};
+ iseq($1,"\xd6","#45605");
+}
+
+{
+ # Regardless of utf8ness any character matches itself when
+ # doing a case insensitive match. See also [perl #36207]
+ for my $o (0..255) {
+ my @ch=(chr($o),chr($o));
+ utf8::upgrade($ch[1]);
+ for my $u_str (0,1) {
+ for my $u_pat (0,1) {
+ ok( $ch[$u_str]=~/\Q$ch[$u_pat]\E/i,
+ "\$c=~/\$c/i : chr($o) : u_str=$u_str u_pat=$u_pat");
+ ok( $ch[$u_str]=~/\Q$ch[$u_pat]\E|xyz/i,
+ "# \$c=~/\$c|xyz/i : chr($o) : u_str=$u_str u_pat=$u_pat");
+ }
+ }
+ }
+}
+{
+ my $a = 3; "" =~ /(??{ $a })/;
+ my $b = $a;
+ iseq($b, $a, "copy of scalar used for postponed subexpression");
+}
+{
+ local $Message = "\$REGMARK in replacement -- Bug #49190";
+ my $_ = "A";
+ s/(*:B)A/$REGMARK/;
+ iseq $_, "B";
+ $_ = "CCCCBAA";
+ s/(*:X)A+|(*:Y)B+|(*:Z)C+/$REGMARK/g;
+ iseq $_, "ZYX";
+}
+if ($::running_as_thread) {
+ for (1..3) {
+ print "not ok $test # TODO & SKIP: croaks when threaded\n";
+ $test++;
+ }
+} else {
+ our @ctl_n=();
+ our @plus=();
+ our $nested_tags;
+ $nested_tags = qr{
+ <
+ (\w+)
+ (?{
+ push @ctl_n,$^N;
+ push @plus,$+;
+ })
+ >
+ (??{$nested_tags})*
+ </\s* \w+ \s*>
+ }x;
+
+ my $match= '<bla><blubb></blubb></bla>' =~ m/^$nested_tags$/;
+ ok($match,'nested construct matches');
+ iseq("@ctl_n","bla blubb",'$^N inside of (?{}) works as expected');
+ iseq("@plus","bla blubb",'$+ inside of (?{}) works as expected');
+}
+
+
+
+
+# Test counter is at bottom of file. Put new tests above here.
+#-------------------------------------------------------------------
# Keep the following tests last -- they may crash perl
+{
+ # RT#19049 / RT#38869
+ my @list = (
+ 'ab cdef', # matches regex
+ ( 'e' x 40000 ) .'ab c' # matches not, but 'ab c' matches part of it
+ );
+ my $y;
+ my $x;
+ foreach (@list) {
+ m/ab(.+)cd/i; # the ignore-case seems to be important
+ $y = $1; # use $1, which might not be from the last match!
+ $x = substr($list[0],$-[0],$+[0]-$-[0]);
+ }
+ iseq($y,' ',
+ 'pattern in a loop, failure should not affect previous success');
+ iseq($x,'ab cd',
+ 'pattern in a loop, failure should not affect previous success');
+}
ok(("a" x (2**15 - 10)) =~ /^()(a|bb)*$/, "Recursive stack cracker: #24274")
or print "# Unexpected outcome: should pass or crash perl\n";
"Regexp /^(??{'(.)'x 100})/ crashes older perls")
or print "# Unexpected outcome: should pass or crash perl\n";
+eval '/\k/';
+ok($@=~/\QSequence \k... not terminated in regex;\E/);
+
{
+ local $Message = "substitution with lookahead (possible segv)";
$_="ns1ns1ns1";
s/ns(?=\d)/ns_/g;
iseq($_,"ns_1ns_1ns_1");
iseq($_,"!Bang!1!Bang!2!Bang!3!Bang!");
}
-# Put new tests above the line, not here.
+# [perl #45337] utf8 + "[a]a{2}" + /$.../ = panic: sv_len_utf8 cache
+
+{
+ local ${^UTF8CACHE} = -1;
+ my $s="[a]a{2}";
+ utf8::upgrade $s;
+ ok("aaa" =~ /$s/, "#45337");
+}
+
+# Put new tests above the dotted line about a page above this comment
+iseq(0+$::test,$::TestCount,"Got the right number of tests!");
+
+} # end of sub pat_tests
# Don't forget to update this!
-BEGIN { print "1..1345\n" };
+BEGIN {
+ $::TestCount = 4019;
+ print "1..$::TestCount\n";
+}
+
+"Truth";