5 @INC = qw(. ../lib); # ../lib needed for test.deparse
11 # Note that t/op/ord.t already tests for chr() <-> ord() rountripping.
15 is(chr(ord("A")), "A");
22 is(chr(-0.1), "\x{FFFD}"); # The U+FFFD Unicode replacement character.
23 is(chr(-1 ), "\x{FFFD}");
24 is(chr(-2 ), "\x{FFFD}");
25 is(chr(-3.0), "\x{FFFD}");
27 use bytes; # Backward compatibility.
28 is(chr(-0.1), "\x00");
31 is(chr(-3.0), "\xFD");
37 no warnings 'utf8'; # avoid surrogate and beyond Unicode warnings
38 join(" ",map{sprintf"%02x",$_}unpack("C*",chr($_[0])));
41 # The following code points are some interesting steps in UTF-8.
42 is(hexes( 0x100), "c4 80");
43 is(hexes( 0x7FF), "df bf");
44 is(hexes( 0x800), "e0 a0 80");
45 is(hexes( 0xFFF), "e0 bf bf");
46 is(hexes( 0x1000), "e1 80 80");
47 is(hexes( 0xCFFF), "ec bf bf");
48 is(hexes( 0xD000), "ed 80 80");
49 is(hexes( 0xD7FF), "ed 9f bf");
50 is(hexes( 0xD800), "ed a0 80"); # not strict utf-8 (surrogate area begin)
51 is(hexes( 0xDFFF), "ed bf bf"); # not strict utf-8 (surrogate area end)
52 is(hexes( 0xE000), "ee 80 80");
53 is(hexes( 0xFFFF), "ef bf bf");
54 is(hexes( 0x10000), "f0 90 80 80");
55 is(hexes( 0x3FFFF), "f0 bf bf bf");
56 is(hexes( 0x40000), "f1 80 80 80");
57 is(hexes( 0xFFFFF), "f3 bf bf bf");
58 is(hexes(0x100000), "f4 80 80 80");
59 is(hexes(0x10FFFF), "f4 8f bf bf"); # Unicode (4.1) last code point
60 is(hexes(0x110000), "f4 90 80 80");
61 is(hexes(0x1FFFFF), "f7 bf bf bf"); # last four byte encoding
62 is(hexes(0x200000), "f8 88 80 80 80");