BEGIN {
chdir 't' if -d 't';
- require './test.pl';
+ require './test.pl'; require './charset_tools.pl';
set_up_inc(qw '../lib ../dist/Math-BigInt/lib');
}
my $no_signedness = $] > 5.009 ? '' :
"Signed/unsigned pack modifiers not available on this perl";
-plan tests => 14704;
+plan tests => 14713;
use strict;
use warnings qw(FATAL all);
use Config;
-my $Is_EBCDIC = (defined $Config{ebcdic} && $Config{ebcdic} eq 'define');
my $Perl = which_perl();
my @valid_errors = (qr/^Invalid type '\w'/);
{
my $sum = 129; # ASCII
- $sum = 103 if $Is_EBCDIC;
+ $sum = 103 if $::IS_EBCDIC;
my $x;
is( ($x = unpack("%32B*", "Now is the time for all good blurfl")), $sum );
# Is this a stupid thing to do on VMS, VOS and other unusual platforms?
skip("-- the IEEE infinity model is unavailable in this configuration.", 1)
- if (($^O eq 'VMS') && !defined($Config{useieee}));
+ if (($^O eq 'VMS') && !defined($Config{useieee}) || !$Config{d_double_has_inf});
skip("-- $^O has serious fp indigestion on w-packed infinities", 1)
if (
if ($^O eq 'vos');
eval { $x = pack 'w', $inf };
- like ($@, qr/^Cannot compress integer/, "Cannot compress integer");
+ like ($@, qr/^Cannot compress Inf/, "Cannot compress infinity");
}
SKIP: {
skip("-- the full range of an IEEE double may not be available in this configuration.", 3)
- if (($^O eq 'VMS') && !defined($Config{useieee}));
+ if (($^O eq 'VMS') && !defined($Config{useieee}) || !$Config{d_double_style_ieee});
skip("-- $^O does not like 2**1023", 3)
if (($^O eq 'ultrix'));
foreach (
['a/a*/a*', '212ab345678901234567','ab3456789012'],
['a/a*/a*', '3012ab345678901234567', 'ab3456789012'],
- ['a/a*/b*', '212ab', $Is_EBCDIC ? '100000010100' : '100001100100'],
+ ['a/a*/b*', '212ab', $::IS_EBCDIC ? '100000010100' : '100001100100'],
)
{
my ($pat, $in, $expect) = @$_;
}
-SKIP: {
- skip("(EBCDIC and) version strings are bad idea", 2) if $Is_EBCDIC;
-
- is("1.20.300.4000", sprintf "%vd", pack("U*",1,20,300,4000));
- is("1.20.300.4000", sprintf "%vd", pack(" U*",1,20,300,4000));
+{
+ is("1.20.300.4000", sprintf "%vd", pack("U*",utf8::native_to_unicode(1),utf8::native_to_unicode(20),300,4000));
+ is("1.20.300.4000", sprintf "%vd", pack(" U*",utf8::native_to_unicode(1),utf8::native_to_unicode(20),300,4000));
}
-isnt(v1.20.300.4000, sprintf "%vd", pack("C0U*",1,20,300,4000));
+isnt(v1.20.300.4000, sprintf "%vd", pack("C0U*",utf8::native_to_unicode(1),utf8::native_to_unicode(20),300,4000));
-my $rslt = $Is_EBCDIC ? "156 67" : "199 162";
+my $rslt = join " ", map { ord } split "", byte_utf8a_to_utf8n("\xc7\xa2");
+# The ASCII UTF-8 of U+1E2 is "\xc7\xa2"
is(join(" ", unpack("U0 C*", chr(0x1e2))), $rslt);
# does pack U create Unicode?
# is unpack U the reverse of pack U for byte string?
is("@{[unpack('U*', pack('U*', 100, 200))]}", "100 200");
-
-SKIP: {
- skip "Not for EBCDIC", 4 if $Is_EBCDIC;
-
+{
# does pack U0C create Unicode?
- is("@{[pack('U0C*', 100, 195, 136)]}", v100.v200);
+ my $cp202 = chr(202);
+ utf8::upgrade $cp202;
+ my @bytes202;
+ { # This is portable across character sets
+ use bytes;
+ @bytes202 = map { ord } split "", $cp202;
+ }
+
+ # This test requires the first number to be invariant; 64 is invariant on
+ # ASCII and EBCDIC.
+ is("@{[pack('U0C*', 64, @bytes202)]}", v64.v202);
# does pack C0U create characters?
- is("@{[pack('C0U*', 100, 200)]}", pack("C*", 100, 195, 136));
+ # The U* is expecting Unicode, so convert to that.
+ is("@{[pack('C0U*', map { utf8::native_to_unicode($_) } 64, 202)]}",
+ pack("C*", 64, @bytes202));
# does unpack U0U on byte data warn?
{
use warnings qw(NONFATAL all);;
- my $bad = pack("U0C", 255);
+ my $bad = pack("U0C", 202);
local $SIG{__WARN__} = sub { $@ = "@_" };
my @null = unpack('U0U', $bad);
- like($@, qr/^Malformed UTF-8 character /);
+ like($@, qr/^Malformed UTF-8 character: /);
}
}
| [Bb] (?{ '101' })
| [Hh] (?{ 'b8' })
| [svnSiIlVNLqQjJ] (?{ 10111 })
- | [FfDd] (?{ 1.36514538e67 })
+ | [FfDd] (?{ 1.36514538e37 })
| [pP] (?{ "try this buffer" })
/x; $^R } @codes;
my @end = (0x12345678, 0x23456781, 0x35465768, 0x15263748);
my @b = unpack "$t X[$t] $t", $p; # Extract, step back, extract again
is(scalar @b, 2 * scalar @a);
$b = "@b";
- $b =~ s/(?:17000+|16999+)\d+(e-45) /17$1 /gi; # stringification is gamble
+ $b =~ s/(?:17000+|16999+)\d+(e-0?45) /17$1 /gi; # stringification is gamble
is($b, "@a @a");
use warnings qw(NONFATAL all);;
is($warning, undef);
is(scalar @b, scalar @a);
$b = "@b";
- $b =~ s/(?:17000+|16999+)\d+(e-45) /17$1 /gi; # stringification is gamble
+ $b =~ s/(?:17000+|16999+)\d+(e-0?45) /17$1 /gi; # stringification is gamble
is($b, "@a");
}
local $SIG{__WARN__} = sub {
$warning = $_[0];
};
- my $out = pack("u99", "foo" x 99);
+
+ # This test is looking for the encoding of the bit pattern "\x66\x6f\x6f",
+ # which is ASCII "foo"
+ my $out = pack("u99", native_to_uni("foo") x 99);
like($warning, qr/Field too wide in 'u' format in pack at /,
"Warn about too wide uuencode");
is($out, ("_" . "9F]O" x 21 . "\n") x 4 . "M" . "9F]O" x 15 . "\n",
my (@y) = unpack("%b10a", "abcd");
is($x[1], $y[1], "checksum advance ok");
- # verify that the checksum is not overflowed with C0
- if (ord('A') == 193) {
- is(unpack("C0%128U", "/bcd"), unpack("U0%128U", "abcd"), "checksum not overflowed");
- } else {
- is(unpack("C0%128U", "abcd"), unpack("U0%128U", "abcd"), "checksum not overflowed");
+ SKIP: {
+ skip("-- non-IEEE float", 1) if !$Config{d_double_style_ieee};
+ # verify that the checksum is not overflowed with C0
+ is(unpack("C0%128U", "abcd"), unpack("U0%128U", "abcd"), "checksum not overflowed");
}
}
+my $U_1FFC_bytes = byte_utf8a_to_utf8n("\341\277\274");
{
# U0 and C0 must be scoped
- my (@x) = unpack("a(U0)U", "b\341\277\274");
+ my (@x) = unpack("a(U0)U", "b$U_1FFC_bytes");
is($x[0], 'b', 'before scope');
is($x[1], 8188, 'after scope');
- is(pack("a(U0)U", "b", 8188), "b\341\277\274");
+ is(pack("a(U0)U", "b", 8188), "b$U_1FFC_bytes");
}
{
# counted length prefixes shouldn't change C0/U0 mode
- # (note the length is actually 0 in this test)
- if (ord('A') == 193) {
- is(join(',', unpack("aU0C/UU", "b\0\341\277\274")), 'b,0');
- is(join(',', unpack("aU0C/CU", "b\0\341\277\274")), 'b,0');
- } else {
- is(join(',', unpack("aC/UU", "b\0\341\277\274")), 'b,8188');
- is(join(',', unpack("aC/CU", "b\0\341\277\274")), 'b,8188');
- is(join(',', unpack("aU0C/UU", "b\0\341\277\274")), 'b,225');
- is(join(',', unpack("aU0C/CU", "b\0\341\277\274")), 'b,225');
- }
+ # (note the length is actually 0 in this test, as the C/ is replaced by C0
+ # due to the \0 in the string)
+ is(join(',', unpack("aC/UU", "b\0$U_1FFC_bytes")), 'b,8188');
+ is(join(',', unpack("aC/CU", "b\0$U_1FFC_bytes")), 'b,8188');
+
+ # The U expects Unicode, so convert from native
+ my $first_byte = utf8::native_to_unicode(ord substr($U_1FFC_bytes, 0, 1));
+
+ is(join(',', unpack("aU0C/UU", "b\0$U_1FFC_bytes")), "b,$first_byte");
+ is(join(',', unpack("aU0C/CU", "b\0$U_1FFC_bytes")), "b,$first_byte");
}
{
is(pack("A*", $high), "\xfeb");
is(pack("Z*", $high), "\xfeb\x00");
- utf8::upgrade($high = "\xc3\xbeb");
- is(pack("U0a2", $high), "\xfe");
- is(pack("U0A2", $high), "\xfe");
- is(pack("U0Z1", $high), "\x00");
- is(pack("U0a3", $high), "\xfeb");
- is(pack("U0A3", $high), "\xfeb");
- is(pack("U0Z3", $high), "\xfe\x00");
- is(pack("U0a6", $high), "\xfeb\x00\x00\x00");
- is(pack("U0A6", $high), "\xfeb ");
- is(pack("U0Z6", $high), "\xfeb\x00\x00\x00");
- is(pack("U0a*", $high), "\xfeb");
- is(pack("U0A*", $high), "\xfeb");
- is(pack("U0Z*", $high), "\xfeb\x00");
+ utf8::upgrade($high = byte_utf8a_to_utf8n("\xc3\xbe") . "b");
+ is(pack("U0a2", $high), uni_to_native("\xfe"));
+ is(pack("U0A2", $high), uni_to_native("\xfe"));
+ is(pack("U0Z1", $high), uni_to_native("\x00"));
+ is(pack("U0a3", $high), uni_to_native("\xfe") . "b");
+ is(pack("U0A3", $high), uni_to_native("\xfe") . "b");
+ is(pack("U0Z3", $high), uni_to_native("\xfe\x00"));
+ is(pack("U0a6", $high), uni_to_native("\xfe") . "b" . uni_to_native("\x00\x00\x00"));
+ is(pack("U0A6", $high), uni_to_native("\xfe") . "b ");
+ is(pack("U0Z6", $high), uni_to_native("\xfe") . "b" . uni_to_native("\x00\x00\x00"));
+ is(pack("U0a*", $high), uni_to_native("\xfe") . "b");
+ is(pack("U0A*", $high), uni_to_native("\xfe") . "b");
+ is(pack("U0Z*", $high), uni_to_native("\xfe") . "b" . uni_to_native("\x00"));
}
{
# pack /
}
{
# unpack("A*", $unicode) strips general unicode spaces
- is(unpack("A*", "ab \n\xa0 \0"), "ab \n\xa0",
+ is(unpack("A*", "ab \n" . uni_to_native("\xa0") . " \0"), "ab \n" . uni_to_native("\xa0"),
'normal A* strip leaves \xa0');
- is(unpack("U0C0A*", "ab \n\xa0 \0"), "ab \n\xa0",
+ is(unpack("U0C0A*", "ab \n" . uni_to_native("\xa0") . " \0"), "ab \n" . uni_to_native("\xa0"),
'normal A* strip leaves \xa0 even if it got upgraded for technical reasons');
is(unpack("A*", pack("a*(U0U)a*", "ab \n", 0xa0, " \0")), "ab",
'upgraded strings A* removes \xa0');
}
{
#50256
- my ($v) = split //, unpack ('(B)*', 'ab');
+ # This test is for the bit pattern "\x61\x62", which is ASCII "ab"
+ my ($v) = split //, unpack ('(B)*', native_to_uni('ab'));
is($v, 0); # Doesn't SEGV :-)
}
{
#90160
is(eval { () = unpack "C0 U*", ""; "ok" }, "ok",
'medial U* on empty string');
+
+package o {
+ use overload
+ '""' => sub { ++$o::str; "42" },
+ '0+' => sub { ++$o::num; 42 };
+}
+is pack("c", bless [], "o"), chr(42), 'overloading called';
+is $o::str, undef, 'pack "c" does not call string overloading';
+is $o::num, 1, 'pack "c" does call num overloading';
+
+#[perl #123874]: argument underflow leads to corrupt length
+eval q{ pack "pi/x" };
+ok(1, "argument underflow did not crash");
+
+{
+ # [perl #126325] pack [hH] with a unicode string
+ # the hex encoders would read past the end of the string, using
+ # invalid source bytes
+ my $twenty_nuls = "\0" x 20;
+ # This is the case that failed
+ is(pack("WH40", 0x100, ""), "\x{100}$twenty_nuls",
+ "check pack H zero fills (utf8 target)");
+ my $up_nul = "\0";
+
+ utf8::upgrade($up_nul);
+ # check the other combinations too
+ is(pack("WH40", 0x100, $up_nul), "\x{100}$twenty_nuls",
+ "check pack H zero fills (utf8 target/source)");
+ is(pack("H40", ""), $twenty_nuls,
+ "check pack H zero fills (utf8 none)");
+ is(pack("H40", $up_nul), $twenty_nuls,
+ "check pack H zero fills (utf8 source)");
+}
+
+SKIP:
+{
+ # [perl #129149] the code below would write one past the end of the output
+ # buffer, only detected by ASAN, not by valgrind
+ $Config{ivsize} >= 8
+ or skip "[perl #129149] need 64-bit for this test", 1;
+ fresh_perl_is(<<'EOS', "ok\n", { stderr => 1 }, "pack W overflow");
+print pack("ucW", "0000", 0, 140737488355327) eq "\$,#`P,```\n\0\x{7fffffffffff}"
+ ? "ok\n" : "not ok\n";
+EOS
+}