--- perl-5.8.8/t/op/lc.t.U27329 2005-11-07 09:22:36.000000000 -0500 +++ perl-5.8.8/t/op/lc.t 2006-06-01 22:02:13.000000000 -0400 @@ -6,7 +6,7 @@ require './test.pl'; } -plan tests => 59; +plan tests => 77; $a = "HELLO.* world"; $b = "hello.* WORLD"; @@ -163,3 +163,38 @@ is($a, v10, "[perl #18857]"); } } + + +# [perl #38619] Bug in lc and uc (interaction between UTF-8, substr, and lc/uc) + +for ("a\x{100}", "xyz\x{100}") { + is(substr(uc($_), 0), uc($_), "[perl #38619] uc"); +} +for ("A\x{100}", "XYZ\x{100}") { + is(substr(lc($_), 0), lc($_), "[perl #38619] lc"); +} +for ("a\x{100}", "ßyz\x{100}") { # ß to Ss (different length) + is(substr(ucfirst($_), 0), ucfirst($_), "[perl #38619] ucfirst"); +} + +# Related to [perl #38619] +# the original report concerns PERL_MAGIC_utf8. +# these cases concern PERL_MAGIC_regex_global. + +for (map { $_ } "a\x{100}", "abc\x{100}", "\x{100}") { + chop; # get ("a", "abc", "") in utf8 + my $return = uc($_) =~ /\G(.?)/g; + my $result = $return ? $1 : "not"; + my $expect = (uc($_) =~ /(.?)/g)[0]; + is($return, 1, "[perl #38619]"); + is($result, $expect, "[perl #38619]"); +} + +for (map { $_ } "A\x{100}", "ABC\x{100}", "\x{100}") { + chop; # get ("A", "ABC", "") in utf8 + my $return = lc($_) =~ /\G(.?)/g; + my $result = $return ? $1 : "not"; + my $expect = (lc($_) =~ /(.?)/g)[0]; + is($return, 1, "[perl #38619]"); + is($result, $expect, "[perl #38619]"); +} --- perl-5.8.8/pp.c.U27329 2006-06-01 21:30:14.000000000 -0400 +++ perl-5.8.8/pp.c 2006-06-01 21:53:37.000000000 -0400 @@ -3447,7 +3447,8 @@ if (slen > ulen) sv_catpvn(TARG, (char*)(s + ulen), slen - ulen); SvUTF8_on(TARG); - SETs(TARG); + sv = TARG; + SETs(sv); } else { s = (U8*)SvPV_force_nomg(sv, slen); @@ -3502,7 +3503,8 @@ if (slen > ulen) sv_catpvn(TARG, (char*)(s + ulen), slen - ulen); SvUTF8_on(TARG); - SETs(TARG); + sv = TARG; + SETs(sv); } else { s = (U8*)SvPV_force_nomg(sv, slen); @@ -3552,7 +3554,8 @@ if (!len) { SvUTF8_off(TARG); /* decontaminate */ sv_setpvn(TARG, "", 0); - SETs(TARG); + sv = TARG; + SETs(sv); } else { STRLEN min = len + 1; @@ -3585,7 +3588,8 @@ *d = '\0'; SvUTF8_on(TARG); SvCUR_set(TARG, d - (U8*)SvPVX_const(TARG)); - SETs(TARG); + sv = TARG; + SETs(sv); } } else { @@ -3636,7 +3640,8 @@ if (!len) { SvUTF8_off(TARG); /* decontaminate */ sv_setpvn(TARG, "", 0); - SETs(TARG); + sv = TARG; + SETs(sv); } else { STRLEN min = len + 1; @@ -3688,7 +3693,8 @@ *d = '\0'; SvUTF8_on(TARG); SvCUR_set(TARG, d - (U8*)SvPVX_const(TARG)); - SETs(TARG); + sv = TARG; + SETs(sv); } } else {