6 unless(grep /blib/, @INC) {
11 $SIG{__WARN__} = sub { push @WARN, @_ };
20 use charnames ':full';
22 print "not " unless "Here\N{EXCLAMATION MARK}?" eq "Here!?";
26 use bytes; # TEST -utf8 can switch utf8 on
28 print "# \$res=$res \$\@='$@'\nnot "
29 if $res = eval <<'EOE'
30 use charnames ":full";
31 "Here: \N{CYRILLIC SMALL LETTER BE}!";
34 or $@ !~ /above 0xFF/;
36 # print "# \$res=$res \$\@='$@'\n";
38 print "# \$res=$res \$\@='$@'\nnot "
39 if $res = eval <<'EOE'
40 use charnames 'cyrillic';
44 or $@ !~ /CYRILLIC CAPITAL LETTER BE.*above 0xFF/;
48 # If octal representation of unicode char is \0xyzt, then the utf8 is \3xy\2zt
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";
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";
68 use charnames ':full';
70 print "not " unless to_bytes("\N{CYRILLIC SMALL LETTER BE}") eq $encoded_be;
73 use charnames qw(cyrillic greek :short);
75 print "not " unless to_bytes("\N{be},\N{alpha},\N{hebrew:bet}")
76 eq "$encoded_be,$encoded_alpha,$encoded_bet";
81 use charnames ':full';
82 print "not " unless "\x{263a}" eq "\N{WHITE SMILING FACE}";
84 print "not " unless length("\x{263a}") == 1;
86 print "not " unless length("\N{WHITE SMILING FACE}") == 1;
88 print "not " unless sprintf("%vx", "\x{263a}") eq "263a";
90 print "not " unless sprintf("%vx", "\N{WHITE SMILING FACE}") eq "263a";
92 print "not " unless sprintf("%vx", "\xFF\N{WHITE SMILING FACE}") eq "ff.263a";
94 print "not " unless sprintf("%vx", "\x{ff}\N{WHITE SMILING FACE}") eq "ff.263a";
99 use charnames qw(:full);
103 my $named = "\N{CUBE ROOT}";
105 print "not " unless ord($x) == ord($named);
110 use charnames qw(:full);
112 print "not " unless "\x{100}\N{CENT SIGN}" eq "\x{100}"."\N{CENT SIGN}";
117 use charnames ':full';
120 unless to_bytes("\N{DESERET SMALL LETTER ENG}") eq $encoded_deseng;
127 no utf8; # naked Latin-1
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;
135 print "ok 16 # Skip: not Latin-1\n";
140 print "not " unless charnames::viacode(0x1234) eq "ETHIOPIC SYLLABLE SEE";
144 print "not " if defined charnames::viacode(0x0590);
150 sprintf("%04X", charnames::vianame("GOTHIC LETTER AHSA")) eq "10330";
154 defined charnames::vianame("NONE SUCH");
159 # check that caching at least hasn't broken anything
161 print "not " unless charnames::viacode(0x1234) eq "ETHIOPIC SYLLABLE SEE";
165 sprintf("%04X", charnames::vianame("GOTHIC LETTER AHSA")) eq "10330";
170 print "not " unless "\N{CHARACTER TABULATION}" eq "\t";
173 print "not " unless "\N{ESCAPE}" eq "\e";
176 print "not " unless "\N{NULL}" eq "\c@";
179 print "not " unless "\N{LINE FEED (LF)}" eq "\n";
182 print "not " unless "\N{LINE FEED}" eq "\n";
185 print "not " unless "\N{LF}" eq "\n";
188 my $nel = ord("A") == 193 ? qr/^(?:\x15|\x25)$/ : qr/^\x85$/;
190 print "not " unless "\N{NEXT LINE (NEL)}" =~ $nel;
193 print "not " unless "\N{NEXT LINE}" =~ $nel;
196 print "not " unless "\N{NEL}" =~ $nel;
199 print "not " unless "\N{BYTE ORDER MARK}" eq chr(0xFEFF);
202 print "not " unless "\N{BOM}" eq chr(0xFEFF);
206 use warnings 'deprecated';
208 print "not " unless "\N{HORIZONTAL TABULATION}" eq "\t";
211 print "not " unless grep { /"HORIZONTAL TABULATION" is deprecated/ } @WARN;
214 no warnings 'deprecated';
216 print "not " unless "\N{VERTICAL TABULATION}" eq "\013";
219 print "not " if grep { /"VERTICAL TABULATION" is deprecated/ } @WARN;
223 print "not " unless charnames::viacode(0xFEFF) eq "ZERO WIDTH NO-BREAK SPACE";
228 print "not " unless ord("\N{BOM}") == 0xFEFF;
232 print "not " unless ord("\N{ZWNJ}") == 0x200C;
235 print "not " unless ord("\N{ZWJ}") == 0x200D;
238 print "not " unless "\N{U+263A}" eq "\N{WHITE SMILING FACE}";
243 0x3093 == charnames::vianame("HIRAGANA LETTER N");
247 0x0397 == charnames::vianame("GREEK CAPITAL LETTER ETA");
251 print "not " if defined charnames::viacode(0x110000);
254 print "not " if grep { /you asked for U+110000/ } @WARN;
257 print "not " unless "NULL" eq charnames::viacode(0);
261 # ---- Alias extensions
263 my $alifile = File::Spec->catfile(File::Spec->updir, qw(lib unicore xyzzy_alias.pl));
268 @prgs = split "\n########\n", <DATA>;
273 my ($code, $exp) = ((split m/\nEXPECT\n/), '$');
274 my ($prog, $fil) = ((split m/\nFILE\n/, $code), "");
275 my $tmpfile = tempfile();
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: $!";
281 open my $ali, "> $alifile" or die "Could not open $alifile: $!";
283 close $ali or die "Could not close $alifile: $!";
285 my $res = runperl( switches => $switch,
286 progfile => $tmpfile,
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
293 $exp =~ s/[\r\n]+$//;
294 my $pfx = ($res =~ s/^PREFIX\n//);
295 my $rexp = qr{^$exp};
296 if ($res =~ s/^SKIPPED\n//) {
299 elsif (($pfx and $res !~ /^\Q$expected/) or
300 (!$pfx and $res !~ $rexp)) {
308 print "ok ", ++$i, "\n";
310 1 while unlink $alifile;
313 # [perl #30409] charnames.pm clobbers default variable
315 eval "use charnames ':full';";
316 print "not " unless $_ eq 'foobar';
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.
325 my $names = do "unicore/Name.pl";
326 print defined $names ? "ok 76\n" : "not ok 76\n";
327 if (ord('A') == 65) { # as on ASCII or UTF-8 machines
328 my $non_ascii = $names =~ tr/\0-\177//c;
329 print $non_ascii ? "not ok 77 # $non_ascii\n" : "ok 77\n";
334 # Verify that charnames propagate to eval("")
335 my $evaltry = eval q[ "Eval: \N{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}" ];
337 print "# $@not ok 78\nnot ok 79\n";
340 print "not " unless $evaltry eq "Eval: \N{LEFT-POINTING DOUBLE ANGLE QUOTATION MARK}";
344 # Verify that db includes the normative NameAliases.txt names
345 print "not " unless "\N{U+1D0C5}" eq "\N{BYZANTINE MUSICAL SYMBOL FTHORA SKLIRON CHROMA VASIS}";
348 # [perl #73174] use of \N{FOO} used to reset %^H
351 use charnames ":full";
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 $res eq "foo-foo-1--2-3" ? "" : "not ",
362 "ok 81 - \$^H{foo} correct after /\\N{bar}/i (res=$res)\n";
367 use charnames ":scoobydoo";
368 "Here: \N{e_ACUTE}!\n";
370 unsupported special ':scoobydoo' in charnames at
372 # wrong type of alias (missing colon)
373 use charnames "alias";
374 "Here: \N{e_ACUTE}!\n";
376 Unknown charname 'e_ACUTE' at
378 # alias without an argument
379 use charnames ":alias";
380 "Here: \N{e_ACUTE}!\n";
382 :alias needs an argument in charnames at
385 use charnames ":alias" => ":full";
386 "Here: \N{e_ACUTE}!\n";
388 :alias cannot use existing pragma :full \(reversed order\?\) at
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";
394 Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
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";
400 Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
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";
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";
412 Unknown charname 'LATIN:e WITH ACUTE' at
414 # alias with hashref to :short OK
415 use charnames ":short", ":alias" => { e_ACUTE => "LATIN:e WITH ACUTE" };
416 "Here: \N{e_ACUTE}!\n";
420 # alias with bad hashref
421 use charnames ":short", ":alias" => "e_ACUTE";
422 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
424 unicore/e_ACUTE_alias.pl cannot be used as alias file for charnames at
426 # alias with arrayref
427 use charnames ":short", ":alias" => [ e_ACUTE => "LATIN:e WITH ACUTE" ];
428 "Here: \N{e_ACUTE}!\n";
430 Only HASH reference supported as argument to :alias at
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";
436 Use of uninitialized value
438 # alias with hashref two aliases
439 use charnames ":short", ":alias" => {
440 e_ACUTE => "LATIN:e WITH ACUTE",
443 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
445 Unknown charname '' at
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",
452 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
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",
461 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
463 Unknown charname 'LATIN SMALL LETTER A WITH ACUT' at
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",
470 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
472 Unknown charname 'LATIN SMALL LETTER A WITH ACUTE' at
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",
479 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
481 Unknown charname 'LATIN:e WITH ACUTE' at
483 # alias with nonexisting file
484 use charnames ":full", ":alias" => "xyzzy";
485 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
487 unicore/xyzzy_alias.pl cannot be used as alias file for charnames at
489 # alias with bad file name
490 use charnames ":full", ":alias" => "xy 7-";
491 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
493 Charnames alias files can only have identifier characters at
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";
499 unicore/perl_alias.pl cannot be used as alias file for charnames at
501 # alias with bad file
502 use charnames ":full", ":alias" => "xyzzy";
503 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
508 unicore/xyzzy_alias.pl did not return a \(valid\) list of alias pairs at
510 # alias with file with empty list
511 use charnames ":full", ":alias" => "xyzzy";
512 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
517 Unknown charname 'e_ACUTE' at
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";
524 ( e_ACUTE => "LATIN:e WITH ACUTE",
525 a_ACUTE => "LATIN:a WITH ACUTE",
528 Unknown charname 'LATIN:e WITH ACUTE' at
530 # alias with :short and file OK
531 use charnames ":short", ":alias" => "xyzzy";
532 "Here: \N{e_ACUTE}\N{a_ACUTE}!\n";
535 ( e_ACUTE => "LATIN:e WITH ACUTE",
536 a_ACUTE => "LATIN:a WITH ACUTE",
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";
546 ( e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
547 a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",
550 Unknown charname 'LATIN SMALL LETTER E WITH ACUTE' at
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";
557 ( e_ACUTE => "LATIN:e WITH ACUTE",
558 a_ACUTE => "LATIN:a WITH ACUTE",
561 Unknown charname 'LATIN:e WITH ACUTE' at
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";
568 ( e_ACUTE => "LATIN SMALL LETTER E WITH ACUTE",
569 a_ACUTE => "LATIN SMALL LETTER A WITH ACUTE",