[perl5.git] / lib / charnames.t

#!./perl

my @WARN;

BEGIN {
    unless(grep /blib/, @INC) {
	chdir 't' if -d 't';
	@INC = '../lib';
	require './test.pl';
    }
    $SIG{__WARN__} = sub { push @WARN, @_ };
}

require File::Spec;

$| = 1;

print "1..81\n";

use charnames ':full';

print "not " unless "Here\N{EXCLAMATION MARK}?" eq "Here!?";
print "ok 1\n";

{
  use bytes;			# TEST -utf8 can switch utf8 on

  print "# \$res=$res \$\@='$@'\nnot "
    if $res = eval <<'EOE'
use charnames ":full";
"Here: \N{CYRILLIC SMALL LETTER BE}!";
1
EOE
      or $@ !~ /above 0xFF/;
  print "ok 2\n";
  # print "# \$res=$res \$\@='$@'\n";

  print "# \$res=$res \$\@='$@'\nnot "
    if $res = eval <<'EOE'
use charnames 'cyrillic';
"Here: \N{Be}!";
1
EOE
      or $@ !~ /CYRILLIC CAPITAL LETTER BE.*above 0xFF/;
  print "ok 3\n";
}

# If octal representation of unicode char is \0xyzt, then the utf8 is \3xy\2zt
if (ord('A') == 65) { # as on ASCII or UTF-8 machines
    $encoded_be = "\320\261";
    $encoded_alpha = "\316\261";
    $encoded_bet = "\327\221";
    $encoded_deseng = "\360\220\221\215";
}
else { # EBCDIC where UTF-EBCDIC may be used (this may be 1047 specific since
       # UTF-EBCDIC is codepage specific)
    $encoded_be = "\270\102\130";
    $encoded_alpha = "\264\130";
    $encoded_bet = "\270\125\130";
    $encoded_deseng = "\336\102\103\124";
}

sub to_bytes {
    unpack"U0a*", shift;
}

{
  use charnames ':full';

  print "not " unless to_bytes("\N{CYRILLIC SMALL LETTER BE}") eq $encoded_be;
  print "ok 4\n";

  use charnames qw(cyrillic greek :short);

  print "not " unless to_bytes("\N{be},\N{alpha},\N{hebrew:bet}")
    eq "$encoded_be,$encoded_alpha,$encoded_bet";
  print "ok 5\n";
}

{
    use charnames ':full';
    print "not " unless "\x{263a}" eq "\N{WHITE SMILING FACE}";
    print "ok 6\n";
    print "not " unless length("\x{263a}") == 1;
    print "ok 7\n";
    print "not " unless length("\N{WHITE SMILING FACE}") == 1;
    print "ok 8\n";
    print "not " unless sprintf("%vx", "\x{263a}") eq "263a";
    print "ok 9\n";
    print "not " unless sprintf("%vx", "\N{WHITE SMILING FACE}") eq "263a";
    print "ok 10\n";
    print "not " unless sprintf("%vx", "\xFF\N{WHITE SMILING FACE}") eq "ff.263a";
    print "ok 11\n";
    print "not " unless sprintf("%vx", "\x{ff}\N{WHITE SMILING FACE}") eq "ff.263a";
    print "ok 12\n";
}

{
   use charnames qw(:full);
   use utf8;

    my $x = "\x{221b}";
    my $named = "\N{CUBE ROOT}";

    print "not " unless ord($x) == ord($named);
    print "ok 13\n";
}

{
   use charnames qw(:full);
   use utf8;
   print "not " unless "\x{100}\N{CENT SIGN}" eq "\x{100}"."\N{CENT SIGN}";
   print "ok 14\n";
}

{
  use charnames ':full';

  print "not "
      unless to_bytes("\N{DESERET SMALL LETTER ENG}") eq $encoded_deseng;
  print "ok 15\n";
}

{
  # 20001114.001

  no utf8; # naked Latin-1

  if (ord("Ä") == 0xc4) { # Try to do this only on Latin-1.
      use charnames ':full';
      my $text = "\N{LATIN CAPITAL LETTER A WITH DIAERESIS}";
      print "not " unless $text eq "\xc4" && ord($text) == 0xc4;
      print "ok 16\n";
  } else {
      print "ok 16 # Skip: not Latin-1\n";
  }
}

{
    print "not " unless charnames::viacode(0x1234) eq "ETHIOPIC SYLLABLE SEE";
    print "ok 17\n";

    # Unused Hebrew.
    print "not " if defined charnames::viacode(0x0590);
    print "ok 18\n";
}

{
    print "not " unless
	sprintf("%04X", charnames::vianame("GOTHIC LETTER AHSA")) eq "10330";
    print "ok 19\n";

    print "not " if
	defined charnames::vianame("NONE SUCH");
    print "ok 20\n";
}

{
    # check that caching at least hasn't broken anything

    print "not " unless charnames::viacode(0x1234) eq "ETHIOPIC SYLLABLE SEE";
    print "ok 21\n";

    print "not " unless
	sprintf("%04X", charnames::vianame("GOTHIC LETTER AHSA")) eq "10330";
    print "ok 22\n";

}

print "not " unless "\N{CHARACTER TABULATION}" eq "\t";
print "ok 23\n";

print "not " unless "\N{ESCAPE}" eq "\e";
print "ok 24\n";

print "not " unless "\N{NULL}" eq "\c@";
print "ok 25\n";

print "not " unless "\N{LINE FEED (LF)}" eq "\n";
print "ok 26\n";

print "not " unless "\N{LINE FEED}" eq "\n";
print "ok 27\n";

print "not " unless "\N{LF}" eq "\n";
print "ok 28\n";

my $nel = ord("A") == 193 ? qr/^(?:\x15|\x25)$/ : qr/^\x85$/;

print "not " unless "\N{NEXT LINE (NEL)}" =~ $nel;
print "ok 29\n";

print "not " unless "\N{NEXT LINE}" =~ $nel;
print "ok 30\n";

print "not " unless "\N{NEL}" =~ $nel;
print "ok 31\n";

print "not " unless "\N{BYTE ORDER MARK}" eq chr(0xFEFF);
print "ok 32\n";

print "not " unless "\N{BOM}" eq chr(0xFEFF);
print "ok 33\n";

{
    use warnings 'deprecated';

    print "not " unless "\N{HORIZONTAL TABULATION}" eq "\t";
    print "ok 34\n";

    print "not " unless grep { /"HORIZONTAL TABULATION" is deprecated/ } @WARN;
    print "ok 35\n";

    no warnings 'deprecated';

    print "not " unless "\N{VERTICAL TABULATION}" eq "\013";
    print "ok 36\n";

    print "not " if grep { /"VERTICAL TABULATION" is deprecated/ } @WARN;
    print "ok 37\n";
}

print "not " unless charnames::viacode(0xFEFF) eq "ZERO WIDTH NO-BREAK SPACE";
print "ok 38\n";

{
    use warnings;
    print "not " unless ord("\N{BOM}") == 0xFEFF;
    print "ok 39\n";
}

print "not " unless ord("\N{ZWNJ}") == 0x200C;
print "ok 40\n";

print "not " unless ord("\N{ZWJ}") == 0x200D;
print "ok 41\n";

print "not " unless "\N{U+263A}" eq "\N{WHITE SMILING FACE}";
print "ok 42\n";

{
    print "not " unless
	0x3093 == charnames::vianame("HIRAGANA LETTER N");
    print "ok 43\n";

    print "not " unless
	0x0397 == charnames::vianame("GREEK CAPITAL LETTER ETA");
    print "ok 44\n";
}

print "not " if defined charnames::viacode(0x110000);
print "ok 45\n";

print "not " if grep { /you asked for U+110000/ } @WARN;
print "ok 46\n";

print "not " unless "NULL" eq charnames::viacode(0);
print "ok 47\n";


# ---- Alias extensions

my $alifile = File::Spec->catfile(File::Spec->updir, qw(lib unicore xyzzy_alias.pl));
my $i = 0;

my @prgs;
{   local $/ = undef;
    @prgs = split "\n########\n", <DATA>;
    }

my $i = 47;
for (@prgs) {
    my ($code, $exp) = ((split m/\nEXPECT\n/), '$');
    my ($prog, $fil) = ((split m/\nFILE\n/, $code), "");
    my $tmpfile = tempfile();
    open my $tmp, "> $tmpfile" or die "Could not open $tmpfile: $!";
    print $tmp $prog, "\n";
    close $tmp or die "Could not close $tmpfile: $!";
    if ($fil) {
	$fil .= "\n";
	open my $ali, "> $alifile" or die "Could not open $alifile: $!";
	print $ali $fil;
	close $ali or die "Could not close $alifile: $!";
	}
    my $res = runperl( switches => $switch, 
                       progfile => $tmpfile,
                       stderr => 1 );
    my $status = $?;
    $res =~ s/[\r\n]+$//;
    $res =~ s/tmp\d+/-/g;			# fake $prog from STDIN
    $res =~ s/\n%[A-Z]+-[SIWEF]-.*$//		# clip off DCL status msg
	if $^O eq "VMS";
    $exp =~ s/[\r\n]+$//;
    my $pfx = ($res =~ s/^PREFIX\n//);
    my $rexp = qr{^$exp};
    if ($res =~ s/^SKIPPED\n//) {
	print "$results\n";
	}
    elsif (($pfx and $res !~ /^\Q$expected/) or
	  (!$pfx and $res !~ $rexp)) {
        print STDERR
	    "PROG:\n$prog\n",
	    "FILE:\n$fil",
	    "EXPECTED:\n$exp\n",
	    "GOT:\n$res\n";
        print "not ";
	}
    print "ok ", ++$i, "\n";
    $fil or next;
    1 while unlink $alifile;
    }

# [perl #30409] charnames.pm clobbers default variable
$_ = 'foobar';
eval "use charnames ':full';";
print "not " unless $_ eq 'foobar';
print "ok 75\n";

# Unicode slowdown noted by Phil Pennock, traced to a bug fix in index
# SADAHIRO Tomoyuki's suggestion is to ensure that the UTF-8ness of both
# arguments are indentical before calling index.
# To do this can take advantage of the fact that unicore/Name.pl is 7 bit
# (or at least should be). So assert that that it's true here.

my $names = do "unicore/Name.pl";
print defined $names ? "ok 76\n" : "not ok 76\n";
if (ord('A') == 65) { # as on ASCII or UTF-8 machines
  my $non_ascii = $names =~ tr/\0-\177//c;
  print $non_ascii ? "not ok 77 # $non_ascii\n" : "ok 77\n";
} else {
  print "ok 77\n";
}

# Verify that charnames propagate to eval("")
my $evaltry = eval q[ "Eval: \N{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}" ];
if ($@) {
    print "# $@not ok 78\nnot ok 79\n";
} else {
    print "ok 78\n";
    print "not " unless $evaltry eq "Eval: \N{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}";
    print "ok 79\n";
}

# Verify that db includes the normative NameAliases.txt names
print "not " unless "\N{U+1D0C5}" eq "\N{BYZANTINE MUSICAL SYMBOL FTHORA SKLIRON CHROMA VASIS}";
print "ok 80\n";

# [perl #73174] use of \N{FOO} used to reset %^H

{
    use charnames ":full";
    my $res;
    BEGIN { $^H{73174} = "foo" }
    BEGIN { $res = ($^H{73174} // "") }
    # forces loading of utf8.pm, which used to reset %^H
    $res .= '-1' if ":" =~ /\N{COLON}/i;
    BEGIN { $res .= '-' . ($^H{73174} // "") }
    $res .= '-' . ($^H{73174} // "");
    $res .= '-2' if ":" =~ /\N{COLON}/;
    $res .= '-3' if ":" =~ /\N{COLON}/i;
    print $res eq "foo-foo-1--2-3" ? "" : "not ",
	"ok 81 - \$^H{foo} correct after /\\N{bar}/i (res=$res)\n";
}

__END__
# unsupported pragma
use charnames ":scoobydoo";
"Here: \N{e_ACUTE}!\n";
EXPECT
unsupported special ':scoobydoo' in charnames at
########
# wrong type of alias (missing colon)
use charnames "alias";
"Here: \N{e_ACUTE}!\n";
EXPECT
Unknown charname 'e_ACUTE' at
########
# alias without an argument
use charnames ":alias";
"Here: \N{e_ACUTE}!\n";
EXPECT
:alias needs an argument in charnames at
########
# reversed sequence
use charnames ":alias" => ":full";
"Here: \N{e_ACUTE}!\n";
EXPECT
:alias cannot use existing pragma :full \(reversed order\?\) at
########
# alias with hashref but no :full
use charnames ":alias" => { e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE" };
"Here: \N{e_ACUTE}!\n";
EXPECT
Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
########
# alias with hashref but with :short
use charnames ":short", ":alias" => { e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE" };
"Here: \N{e_ACUTE}!\n";
EXPECT
Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
########
# alias with hashref to :full OK
use charnames ":full", ":alias" => { e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE" };
"Here: \N{e_ACUTE}!\n";
EXPECT
$
########
# alias with hashref to :short but using :full
use charnames ":full", ":alias" => { e_ACUTE => "LATIN:e WITH ACUTE" };
"Here: \N{e_ACUTE}!\n";
EXPECT
Unknown charname 'LATIN:e WITH ACUTE' at
########
# alias with hashref to :short OK
use charnames ":short", ":alias" => { e_ACUTE => "LATIN:e WITH ACUTE" };
"Here: \N{e_ACUTE}!\n";
EXPECT
$
########
# alias with bad hashref
use charnames ":short", ":alias" => "e_ACUTE";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
unicore/e_ACUTE_alias.pl cannot be used as alias file for charnames at
########
# alias with arrayref
use charnames ":short", ":alias" => [ e_ACUTE => "LATIN:e WITH ACUTE" ];
"Here: \N{e_ACUTE}!\n";
EXPECT
Only HASH reference supported as argument to :alias at
########
# alias with bad hashref
use charnames ":short", ":alias" => { e_ACUTE => "LATIN:e WITH ACUTE", "a_ACUTE" };
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
Use of uninitialized value
########
# alias with hashref two aliases
use charnames ":short", ":alias" => {
    e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "",
    };
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
Unknown charname '' at
########
# alias with hashref two aliases
use charnames ":short", ":alias" => {
    e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "LATIN:a WITH ACUTE",
    };
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
$
########
# alias with hashref using mixed aliasses
use charnames ":short", ":alias" => {
    e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "LATIN SMALL LETTER A WITH ACUT",
    };
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
Unknown charname 'LATIN SMALL LETTER A WITH ACUT' at
########
# alias with hashref using mixed aliasses
use charnames ":short", ":alias" => {
    e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
    };
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
Unknown charname 'LATIN SMALL LETTER A WITH ACUTE' at
########
# alias with hashref using mixed aliasses
use charnames ":full", ":alias" => {
    e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
    };
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
Unknown charname 'LATIN:e WITH ACUTE' at
########
# alias with nonexisting file
use charnames ":full", ":alias" => "xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
unicore/xyzzy_alias.pl cannot be used as alias file for charnames at
########
# alias with bad file name
use charnames ":full", ":alias" => "xy 7-";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
Charnames alias files can only have identifier characters at
########
# alias with non_absolute (existing) file name (which it should /not/ use)
use charnames ":full", ":alias" => "perl";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
EXPECT
unicore/perl_alias.pl cannot be used as alias file for charnames at
########
# alias with bad file
use charnames ":full", ":alias" => "xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
FILE
#!perl
0;
EXPECT
unicore/xyzzy_alias.pl did not return a \(valid\) list of alias pairs at
########
# alias with file with empty list
use charnames ":full", ":alias" => "xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
FILE
#!perl
();
EXPECT
Unknown charname 'e_ACUTE' at
########
# alias with file OK but file has :short aliasses
use charnames ":full", ":alias" => "xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
FILE
#!perl
(   e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "LATIN:a WITH ACUTE",
    );
EXPECT
Unknown charname 'LATIN:e WITH ACUTE' at
########
# alias with :short and file OK
use charnames ":short", ":alias" => "xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
FILE
#!perl
(   e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "LATIN:a WITH ACUTE",
    );
EXPECT
$
########
# alias with :short and file OK has :long aliasses
use charnames ":short", ":alias" => "xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
FILE
#!perl
(   e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
    a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
    );
EXPECT
Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
########
# alias with file implicit :full but file has :short aliasses
use charnames ":alias" => ":xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
FILE
#!perl
(   e_ACUTE => "LATIN:e WITH ACUTE",
    a_ACUTE => "LATIN:a WITH ACUTE",
    );
EXPECT
Unknown charname 'LATIN:e WITH ACUTE' at
########
# alias with file implicit :full and file has :long aliasses
use charnames ":alias" => ":xyzzy";
"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
FILE
#!perl
(   e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
    a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
    );
EXPECT
$
Commit	Line	Data
423cee85 JH	1	#!./perl
423cee85 JH	2
52ea3e69 JH	3	my @WARN;
52ea3e69 JH	4
423cee85 JH	5	BEGIN {
	6	unless(grep /blib/, @INC) {
	7	chdir 't' if -d 't';
20822f61	8	@INC = '../lib';
143a3e5e	9	require './test.pl';
423cee85	10	}
52ea3e69	11	$SIG{__WARN__} = sub { push @WARN, @_ };
423cee85 JH	12	}
423cee85 JH	13
63c6dcc1 JH	14	require File::Spec;
63c6dcc1 JH	15
423cee85	16	$\| = 1;
822ebcc8	17
e10d7780	18	print "1..81\n";
423cee85 JH	19
	20	use charnames ':full';
	21
93979888	22	print "not " unless "Here\N{EXCLAMATION MARK}?" eq "Here!?";
423cee85 JH	23	print "ok 1\n";
423cee85 JH	24
c82a54e6	25	{
5d9a6404	26	use bytes; # TEST -utf8 can switch utf8 on
c82a54e6 IZ	27
	28	print "# \$res=$res \$\@='$@'\nnot "
	29	if $res = eval <<'EOE'
423cee85	30	use charnames ":full";
4a2d328f	31	"Here: \N{CYRILLIC SMALL LETTER BE}!";
423cee85 JH	32	1
423cee85 JH	33	EOE
c82a54e6 IZ	34	or $@ !~ /above 0xFF/;
	35	print "ok 2\n";
	36	# print "# \$res=$res \$\@='$@'\n";
423cee85	37
c82a54e6 IZ	38	print "# \$res=$res \$\@='$@'\nnot "
c82a54e6 IZ	39	if $res = eval <<'EOE'
423cee85	40	use charnames 'cyrillic';
4a2d328f	41	"Here: \N{Be}!";
423cee85 JH	42	1
423cee85 JH	43	EOE
c82a54e6 IZ	44	or $@ !~ /CYRILLIC CAPITAL LETTER BE.*above 0xFF/;
	45	print "ok 3\n";
	46	}
423cee85 JH	47
423cee85 JH	48	# If octal representation of unicode char is \0xyzt, then the utf8 is \3xy\2zt
210db7fc PP	49	if (ord('A') == 65) { # as on ASCII or UTF-8 machines
	50	$encoded_be = "\320\261";
	51	$encoded_alpha = "\316\261";
	52	$encoded_bet = "\327\221";
	53	$encoded_deseng = "\360\220\221\215";
	54	}
	55	else { # EBCDIC where UTF-EBCDIC may be used (this may be 1047 specific since
	56	# UTF-EBCDIC is codepage specific)
	57	$encoded_be = "\270\102\130";
	58	$encoded_alpha = "\264\130";
	59	$encoded_bet = "\270\125\130";
	60	$encoded_deseng = "\336\102\103\124";
	61	}
c5cc3500 GS	62
c5cc3500 GS	63	sub to_bytes {
f337b084	64	unpack"U0a*", shift;
c5cc3500 GS	65	}
c5cc3500 GS	66
423cee85 JH	67	{
423cee85 JH	68	use charnames ':full';
423cee85	69
c5cc3500	70	print "not " unless to_bytes("\N{CYRILLIC SMALL LETTER BE}") eq $encoded_be;
423cee85 JH	71	print "ok 4\n";
	72
	73	use charnames qw(cyrillic greek :short);
	74
c5cc3500	75	print "not " unless to_bytes("\N{be},\N{alpha},\N{hebrew:bet}")
423cee85 JH	76	eq "$encoded_be,$encoded_alpha,$encoded_bet";
	77	print "ok 5\n";
	78	}
e1992b6d GS	79
	80	{
	81	use charnames ':full';
	82	print "not " unless "\x{263a}" eq "\N{WHITE SMILING FACE}";
	83	print "ok 6\n";
	84	print "not " unless length("\x{263a}") == 1;
	85	print "ok 7\n";
	86	print "not " unless length("\N{WHITE SMILING FACE}") == 1;
	87	print "ok 8\n";
	88	print "not " unless sprintf("%vx", "\x{263a}") eq "263a";
	89	print "ok 9\n";
	90	print "not " unless sprintf("%vx", "\N{WHITE SMILING FACE}") eq "263a";
	91	print "ok 10\n";
f08d6ad9 GS	92	print "not " unless sprintf("%vx", "\xFF\N{WHITE SMILING FACE}") eq "ff.263a";
	93	print "ok 11\n";
	94	print "not " unless sprintf("%vx", "\x{ff}\N{WHITE SMILING FACE}") eq "ff.263a";
	95	print "ok 12\n";
e1992b6d	96	}
c00525d4 SP	97
	98	{
	99	use charnames qw(:full);
55eda711	100	use utf8;
51cf30b6	101
c00525d4 SP	102	my $x = "\x{221b}";
	103	my $named = "\N{CUBE ROOT}";
	104
	105	print "not " unless ord($x) == ord($named);
	106	print "ok 13\n";
	107	}
	108
f9a63242 JH	109	{
f9a63242 JH	110	use charnames qw(:full);
55eda711	111	use utf8;
f9a63242 JH	112	print "not " unless "\x{100}\N{CENT SIGN}" eq "\x{100}"."\N{CENT SIGN}";
	113	print "ok 14\n";
	114	}
	115
b896c7a5 A	116	{
	117	use charnames ':full';
	118
	119	print "not "
	120	unless to_bytes("\N{DESERET SMALL LETTER ENG}") eq $encoded_deseng;
	121	print "ok 15\n";
4765795a	122	}
b896c7a5	123
4765795a	124	{
51cf30b6	125	# 20001114.001
4765795a	126
4c53e876	127	no utf8; # naked Latin-1
3ba0e062	128
4765795a JH	129	if (ord("Ä") == 0xc4) { # Try to do this only on Latin-1.
	130	use charnames ':full';
	131	my $text = "\N{LATIN CAPITAL LETTER A WITH DIAERESIS}";
	132	print "not " unless $text eq "\xc4" && ord($text) == 0xc4;
	133	print "ok 16\n";
	134	} else {
	135	print "ok 16 # Skip: not Latin-1\n";
	136	}
b896c7a5 A	137	}
b896c7a5 A	138
daf0d493 JH	139	{
	140	print "not " unless charnames::viacode(0x1234) eq "ETHIOPIC SYLLABLE SEE";
	141	print "ok 17\n";
	142
a23c04e4	143	# Unused Hebrew.
11881cb4	144	print "not " if defined charnames::viacode(0x0590);
daf0d493 JH	145	print "ok 18\n";
	146	}
	147
	148	{
	149	print "not " unless
51b0dbc4	150	sprintf("%04X", charnames::vianame("GOTHIC LETTER AHSA")) eq "10330";
daf0d493 JH	151	print "ok 19\n";
	152
	153	print "not " if
	154	defined charnames::vianame("NONE SUCH");
	155	print "ok 20\n";
	156	}
4e2cda5d JH	157
	158	{
	159	# check that caching at least hasn't broken anything
	160
	161	print "not " unless charnames::viacode(0x1234) eq "ETHIOPIC SYLLABLE SEE";
	162	print "ok 21\n";
	163
	164	print "not " unless
51b0dbc4	165	sprintf("%04X", charnames::vianame("GOTHIC LETTER AHSA")) eq "10330";
4e2cda5d JH	166	print "ok 22\n";
	167
	168	}
301a3cda	169
822ebcc8	170	print "not " unless "\N{CHARACTER TABULATION}" eq "\t";
301a3cda JH	171	print "ok 23\n";
	172
	173	print "not " unless "\N{ESCAPE}" eq "\e";
	174	print "ok 24\n";
	175
	176	print "not " unless "\N{NULL}" eq "\c@";
	177	print "ok 25\n";
	178
7b903762 RGS	179	print "not " unless "\N{LINE FEED (LF)}" eq "\n";
7b903762 RGS	180	print "ok 26\n";
eb380778	181
7b903762 RGS	182	print "not " unless "\N{LINE FEED}" eq "\n";
7b903762 RGS	183	print "ok 27\n";
52ea3e69	184
7b903762 RGS	185	print "not " unless "\N{LF}" eq "\n";
7b903762 RGS	186	print "ok 28\n";
52ea3e69	187
a2e77dd4 JH	188	my $nel = ord("A") == 193 ? qr/^(?:\x15\|\x25)$/ : qr/^\x85$/;
	189
	190	print "not " unless "\N{NEXT LINE (NEL)}" =~ $nel;
52ea3e69 JH	191	print "ok 29\n";
52ea3e69 JH	192
a2e77dd4	193	print "not " unless "\N{NEXT LINE}" =~ $nel;
52ea3e69 JH	194	print "ok 30\n";
52ea3e69 JH	195
a2e77dd4	196	print "not " unless "\N{NEL}" =~ $nel;
51e9e896 PN	197	print "ok 31\n";
51e9e896 PN	198
274085e3	199	print "not " unless "\N{BYTE ORDER MARK}" eq chr(0xFEFF);
51e9e896 PN	200	print "ok 32\n";
51e9e896 PN	201
d7d589a8	202	print "not " unless "\N{BOM}" eq chr(0xFEFF);
51e9e896 PN	203	print "ok 33\n";
51e9e896 PN	204
52ea3e69 JH	205	{
	206	use warnings 'deprecated';
	207
	208	print "not " unless "\N{HORIZONTAL TABULATION}" eq "\t";
51e9e896	209	print "ok 34\n";
52ea3e69 JH	210
52ea3e69 JH	211	print "not " unless grep { /"HORIZONTAL TABULATION" is deprecated/ } @WARN;
51e9e896	212	print "ok 35\n";
822ebcc8	213
52ea3e69 JH	214	no warnings 'deprecated';
	215
	216	print "not " unless "\N{VERTICAL TABULATION}" eq "\013";
51e9e896	217	print "ok 36\n";
52ea3e69 JH	218
52ea3e69 JH	219	print "not " if grep { /"VERTICAL TABULATION" is deprecated/ } @WARN;
51e9e896	220	print "ok 37\n";
52ea3e69	221	}
822ebcc8	222
274085e3	223	print "not " unless charnames::viacode(0xFEFF) eq "ZERO WIDTH NO-BREAK SPACE";
a23c04e4 JH	224	print "ok 38\n";
a23c04e4 JH	225
872c91ae JH	226	{
	227	use warnings;
	228	print "not " unless ord("\N{BOM}") == 0xFEFF;
	229	print "ok 39\n";
	230	}
	231
24b5d5cc JH	232	print "not " unless ord("\N{ZWNJ}") == 0x200C;
	233	print "ok 40\n";
	234
	235	print "not " unless ord("\N{ZWJ}") == 0x200D;
	236	print "ok 41\n";
dbc0d4f2 JH	237
	238	print "not " unless "\N{U+263A}" eq "\N{WHITE SMILING FACE}";
	239	print "ok 42\n";
	240
51b0dbc4 TS	241	{
	242	print "not " unless
	243	0x3093 == charnames::vianame("HIRAGANA LETTER N");
	244	print "ok 43\n";
	245
	246	print "not " unless
	247	0x0397 == charnames::vianame("GREEK CAPITAL LETTER ETA");
	248	print "ok 44\n";
	249	}
	250
00d835f2	251	print "not " if defined charnames::viacode(0x110000);
51b0dbc4	252	print "ok 45\n";
00d835f2 JH	253
00d835f2 JH	254	print "not " if grep { /you asked for U+110000/ } @WARN;
51b0dbc4	255	print "ok 46\n";
35c0985d	256
e10d7780 KW	257	print "not " unless "NULL" eq charnames::viacode(0);
	258	print "ok 47\n";
	259
35c0985d MB	260
	261	# ---- Alias extensions
	262
63c6dcc1	263	my $alifile = File::Spec->catfile(File::Spec->updir, qw(lib unicore xyzzy_alias.pl));
35c0985d	264	my $i = 0;
35c0985d MB	265
	266	my @prgs;
	267	{ local $/ = undef;
	268	@prgs = split "\n########\n", <DATA>;
	269	}
	270
e10d7780	271	my $i = 47;
35c0985d MB	272	for (@prgs) {
	273	my ($code, $exp) = ((split m/\nEXPECT\n/), '$');
	274	my ($prog, $fil) = ((split m/\nFILE\n/, $code), "");
f89542f7	275	my $tmpfile = tempfile();
35c0985d MB	276	open my $tmp, "> $tmpfile" or die "Could not open $tmpfile: $!";
	277	print $tmp $prog, "\n";
	278	close $tmp or die "Could not close $tmpfile: $!";
	279	if ($fil) {
	280	$fil .= "\n";
	281	open my $ali, "> $alifile" or die "Could not open $alifile: $!";
	282	print $ali $fil;
	283	close $ali or die "Could not close $alifile: $!";
	284	}
143a3e5e CB	285	my $res = runperl( switches => $switch,
	286	progfile => $tmpfile,
	287	stderr => 1 );
35c0985d MB	288	my $status = $?;
	289	$res =~ s/[\r\n]+$//;
	290	$res =~ s/tmp\d+/-/g; # fake $prog from STDIN
	291	$res =~ s/\n%[A-Z]+-[SIWEF]-.*$// # clip off DCL status msg
	292	if $^O eq "VMS";
	293	$exp =~ s/[\r\n]+$//;
35c0985d MB	294	my $pfx = ($res =~ s/^PREFIX\n//);
	295	my $rexp = qr{^$exp};
	296	if ($res =~ s/^SKIPPED\n//) {
	297	print "$results\n";
	298	}
	299	elsif (($pfx and $res !~ /^\Q$expected/) or
	300	(!$pfx and $res !~ $rexp)) {
	301	print STDERR
	302	"PROG:\n$prog\n",
	303	"FILE:\n$fil",
	304	"EXPECTED:\n$exp\n",
	305	"GOT:\n$res\n";
	306	print "not ";
	307	}
	308	print "ok ", ++$i, "\n";
35c0985d MB	309	$fil or next;
	310	1 while unlink $alifile;
	311	}
	312
e5c3f898 MG	313	# [perl #30409] charnames.pm clobbers default variable
	314	$_ = 'foobar';
	315	eval "use charnames ':full';";
	316	print "not " unless $_ eq 'foobar';
e10d7780	317	print "ok 75\n";
e5c3f898	318
c776535e NC	319	# Unicode slowdown noted by Phil Pennock, traced to a bug fix in index
	320	# SADAHIRO Tomoyuki's suggestion is to ensure that the UTF-8ness of both
	321	# arguments are indentical before calling index.
	322	# To do this can take advantage of the fact that unicore/Name.pl is 7 bit
	323	# (or at least should be). So assert that that it's true here.
	324
	325	my $names = do "unicore/Name.pl";
e10d7780	326	print defined $names ? "ok 76\n" : "not ok 76\n";
c776535e NC	327	if (ord('A') == 65) { # as on ASCII or UTF-8 machines
c776535e NC	328	my $non_ascii = $names =~ tr/\0-\177//c;
e10d7780	329	print $non_ascii ? "not ok 77 # $non_ascii\n" : "ok 77\n";
c776535e	330	} else {
e10d7780	331	print "ok 77\n";
c776535e NC	332	}
c776535e NC	333
eb915052 RGS	334	# Verify that charnames propagate to eval("")
	335	my $evaltry = eval q[ "Eval: \N{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}" ];
	336	if ($@) {
e10d7780	337	print "# $@not ok 78\nnot ok 79\n";
eb915052	338	} else {
eb915052	339	print "ok 78\n";
e10d7780 KW	340	print "not " unless $evaltry eq "Eval: \N{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}";
e10d7780 KW	341	print "ok 79\n";
eb915052	342	}
c776535e	343
ae6979a8 KW	344	# Verify that db includes the normative NameAliases.txt names
ae6979a8 KW	345	print "not " unless "\N{U+1D0C5}" eq "\N{BYZANTINE MUSICAL SYMBOL FTHORA SKLIRON CHROMA VASIS}";
e10d7780	346	print "ok 80\n";
ae6979a8	347
ec34a119 DM	348	# [perl #73174] use of \N{FOO} used to reset %^H
	349
	350	{
	351	use charnames ":full";
	352	my $res;
	353	BEGIN { $^H{73174} = "foo" }
	354	BEGIN { $res = ($^H{73174} // "") }
	355	# forces loading of utf8.pm, which used to reset %^H
	356	$res .= '-1' if ":" =~ /\N{COLON}/i;
	357	BEGIN { $res .= '-' . ($^H{73174} // "") }
	358	$res .= '-' . ($^H{73174} // "");
	359	$res .= '-2' if ":" =~ /\N{COLON}/;
	360	$res .= '-3' if ":" =~ /\N{COLON}/i;
	361	print $res eq "foo-foo-1--2-3" ? "" : "not ",
e10d7780	362	"ok 81 - \$^H{foo} correct after /\\N{bar}/i (res=$res)\n";
ec34a119 DM	363	}
ec34a119 DM	364
35c0985d	365	__END__
51cf30b6 MB	366	# unsupported pragma
	367	use charnames ":scoobydoo";
	368	"Here: \N{e_ACUTE}!\n";
	369	EXPECT
	370	unsupported special ':scoobydoo' in charnames at
	371	########
35c0985d MB	372	# wrong type of alias (missing colon)
	373	use charnames "alias";
	374	"Here: \N{e_ACUTE}!\n";
	375	EXPECT
51cf30b6	376	Unknown charname 'e_ACUTE' at
35c0985d MB	377	########
	378	# alias without an argument
	379	use charnames ":alias";
	380	"Here: \N{e_ACUTE}!\n";
	381	EXPECT
51cf30b6 MB	382	:alias needs an argument in charnames at
	383	########
	384	# reversed sequence
	385	use charnames ":alias" => ":full";
	386	"Here: \N{e_ACUTE}!\n";
	387	EXPECT
	388	:alias cannot use existing pragma :full \(reversed order\?\) at
35c0985d MB	389	########
	390	# alias with hashref but no :full
	391	use charnames ":alias" => { e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE" };
	392	"Here: \N{e_ACUTE}!\n";
	393	EXPECT
	394	Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
	395	########
	396	# alias with hashref but with :short
	397	use charnames ":short", ":alias" => { e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE" };
	398	"Here: \N{e_ACUTE}!\n";
	399	EXPECT
	400	Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
	401	########
	402	# alias with hashref to :full OK
	403	use charnames ":full", ":alias" => { e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE" };
	404	"Here: \N{e_ACUTE}!\n";
	405	EXPECT
	406	$
	407	########
	408	# alias with hashref to :short but using :full
	409	use charnames ":full", ":alias" => { e_ACUTE => "LATIN:e WITH ACUTE" };
	410	"Here: \N{e_ACUTE}!\n";
	411	EXPECT
	412	Unknown charname 'LATIN:e WITH ACUTE' at
	413	########
	414	# alias with hashref to :short OK
	415	use charnames ":short", ":alias" => { e_ACUTE => "LATIN:e WITH ACUTE" };
	416	"Here: \N{e_ACUTE}!\n";
	417	EXPECT
	418	$
	419	########
	420	# alias with bad hashref
	421	use charnames ":short", ":alias" => "e_ACUTE";
	422	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	423	EXPECT
51cf30b6	424	unicore/e_ACUTE_alias.pl cannot be used as alias file for charnames at
35c0985d MB	425	########
	426	# alias with arrayref
	427	use charnames ":short", ":alias" => [ e_ACUTE => "LATIN:e WITH ACUTE" ];
	428	"Here: \N{e_ACUTE}!\n";
	429	EXPECT
	430	Only HASH reference supported as argument to :alias at
	431	########
	432	# alias with bad hashref
	433	use charnames ":short", ":alias" => { e_ACUTE => "LATIN:e WITH ACUTE", "a_ACUTE" };
	434	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	435	EXPECT
29489e7c	436	Use of uninitialized value
35c0985d MB	437	########
	438	# alias with hashref two aliases
	439	use charnames ":short", ":alias" => {
	440	e_ACUTE => "LATIN:e WITH ACUTE",
	441	a_ACUTE => "",
	442	};
	443	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	444	EXPECT
	445	Unknown charname '' at
	446	########
	447	# alias with hashref two aliases
	448	use charnames ":short", ":alias" => {
	449	e_ACUTE => "LATIN:e WITH ACUTE",
	450	a_ACUTE => "LATIN:a WITH ACUTE",
	451	};
	452	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	453	EXPECT
	454	$
	455	########
	456	# alias with hashref using mixed aliasses
	457	use charnames ":short", ":alias" => {
	458	e_ACUTE => "LATIN:e WITH ACUTE",
	459	a_ACUTE => "LATIN SMALL LETTER A WITH ACUT",
	460	};
	461	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	462	EXPECT
	463	Unknown charname 'LATIN SMALL LETTER A WITH ACUT' at
	464	########
	465	# alias with hashref using mixed aliasses
	466	use charnames ":short", ":alias" => {
	467	e_ACUTE => "LATIN:e WITH ACUTE",
	468	a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
	469	};
	470	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	471	EXPECT
	472	Unknown charname 'LATIN SMALL LETTER A WITH ACUTE' at
	473	########
	474	# alias with hashref using mixed aliasses
	475	use charnames ":full", ":alias" => {
	476	e_ACUTE => "LATIN:e WITH ACUTE",
	477	a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
	478	};
	479	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	480	EXPECT
	481	Unknown charname 'LATIN:e WITH ACUTE' at
	482	########
	483	# alias with nonexisting file
	484	use charnames ":full", ":alias" => "xyzzy";
	485	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	486	EXPECT
51cf30b6 MB	487	unicore/xyzzy_alias.pl cannot be used as alias file for charnames at
	488	########
	489	# alias with bad file name
	490	use charnames ":full", ":alias" => "xy 7-";
	491	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	492	EXPECT
	493	Charnames alias files can only have identifier characters at
	494	########
	495	# alias with non_absolute (existing) file name (which it should /not/ use)
	496	use charnames ":full", ":alias" => "perl";
	497	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	498	EXPECT
	499	unicore/perl_alias.pl cannot be used as alias file for charnames at
35c0985d MB	500	########
	501	# alias with bad file
	502	use charnames ":full", ":alias" => "xyzzy";
	503	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	504	FILE
	505	#!perl
	506	0;
	507	EXPECT
51cf30b6	508	unicore/xyzzy_alias.pl did not return a \(valid\) list of alias pairs at
35c0985d MB	509	########
	510	# alias with file with empty list
	511	use charnames ":full", ":alias" => "xyzzy";
	512	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	513	FILE
	514	#!perl
	515	();
	516	EXPECT
	517	Unknown charname 'e_ACUTE' at
	518	########
	519	# alias with file OK but file has :short aliasses
	520	use charnames ":full", ":alias" => "xyzzy";
	521	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	522	FILE
	523	#!perl
	524	( e_ACUTE => "LATIN:e WITH ACUTE",
	525	a_ACUTE => "LATIN:a WITH ACUTE",
	526	);
	527	EXPECT
	528	Unknown charname 'LATIN:e WITH ACUTE' at
	529	########
	530	# alias with :short and file OK
	531	use charnames ":short", ":alias" => "xyzzy";
	532	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	533	FILE
	534	#!perl
	535	( e_ACUTE => "LATIN:e WITH ACUTE",
	536	a_ACUTE => "LATIN:a WITH ACUTE",
	537	);
	538	EXPECT
	539	$
	540	########
	541	# alias with :short and file OK has :long aliasses
	542	use charnames ":short", ":alias" => "xyzzy";
	543	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	544	FILE
	545	#!perl
	546	( e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
	547	a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
	548	);
	549	EXPECT
	550	Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
	551	########
	552	# alias with file implicit :full but file has :short aliasses
	553	use charnames ":alias" => ":xyzzy";
	554	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	555	FILE
	556	#!perl
	557	( e_ACUTE => "LATIN:e WITH ACUTE",
	558	a_ACUTE => "LATIN:a WITH ACUTE",
	559	);
	560	EXPECT
	561	Unknown charname 'LATIN:e WITH ACUTE' at
	562	########
	563	# alias with file implicit :full and file has :long aliasses
	564	use charnames ":alias" => ":xyzzy";
	565	"Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
	566	FILE
	567	#!perl
	568	( e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
	569	a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
	570	);
	571	EXPECT
	572	$