no utf8; # needed for use utf8 not griping about the raw octets
-require "./test.pl";
+BEGIN { require "./test.pl"; }
-plan(tests => 52);
+plan(tests => 55);
$| = 1;
-open(F,"+>:utf8",'a');
+my $a_file = tempfile();
+
+open(F,"+>:utf8",$a_file);
print F chr(0x100).'£';
-ok( tell(F) == 4, tell(F) );
+cmp_ok( tell(F), '==', 4, tell(F) );
print F "\n";
-ok( tell(F) >= 5, tell(F) );
+cmp_ok( tell(F), '>=', 5, tell(F) );
seek(F,0,0);
-ok( getc(F) eq chr(0x100) );
-ok( getc(F) eq "£" );
-ok( getc(F) eq "\n" );
+is( getc(F), chr(0x100) );
+is( getc(F), "£" );
+is( getc(F), "\n" );
seek(F,0,0);
binmode(F,":bytes");
my $chr = chr(0xc4);
-if (ord('A') == 193) { $chr = chr(0x8c); } # EBCDIC
-ok( getc(F) eq $chr );
+if (ord($a_file) == 193) { $chr = chr(0x8c); } # EBCDIC
+is( getc(F), $chr );
$chr = chr(0x80);
-if (ord('A') == 193) { $chr = chr(0x41); } # EBCDIC
-ok( getc(F) eq $chr );
+if (ord($a_file) == 193) { $chr = chr(0x41); } # EBCDIC
+is( getc(F), $chr );
$chr = chr(0xc2);
-if (ord('A') == 193) { $chr = chr(0x80); } # EBCDIC
-ok( getc(F) eq $chr );
+if (ord($a_file) == 193) { $chr = chr(0x80); } # EBCDIC
+is( getc(F), $chr );
$chr = chr(0xa3);
-if (ord('A') == 193) { $chr = chr(0x44); } # EBCDIC
-ok( getc(F) eq $chr );
-ok( getc(F) eq "\n" );
+if (ord($a_file) == 193) { $chr = chr(0x44); } # EBCDIC
+is( getc(F), $chr );
+is( getc(F), "\n" );
seek(F,0,0);
binmode(F,":utf8");
-ok( scalar(<F>) eq "\x{100}£\n" );
+is( scalar(<F>), "\x{100}£\n" );
seek(F,0,0);
$buf = chr(0x200);
$count = read(F,$buf,2,1);
-ok( $count == 2 );
-ok( $buf eq "\x{200}\x{100}£" );
+cmp_ok( $count, '==', 2 );
+is( $buf, "\x{200}\x{100}£" );
close(F);
{
$a = chr(300); # This *is* UTF-encoded
$b = chr(130); # This is not.
- open F, ">:utf8", 'a' or die $!;
+ open F, ">:utf8", $a_file or die $!;
print F $a,"\n";
close F;
- open F, "<:utf8", 'a' or die $!;
+ open F, "<:utf8", $a_file or die $!;
$x = <F>;
chomp($x);
- ok( $x eq chr(300) );
+ is( $x, chr(300) );
- open F, "a" or die $!; # Not UTF
+ open F, $a_file or die $!; # Not UTF
binmode(F, ":bytes");
$x = <F>;
chomp($x);
$chr = chr(196).chr(172);
- if (ord('A') == 193) { $chr = chr(141).chr(83); } # EBCDIC
- ok( $x eq $chr );
+ if (ord($a_file) == 193) { $chr = chr(141).chr(83); } # EBCDIC
+ is( $x, $chr );
close F;
- open F, ">:utf8", 'a' or die $!;
+ open F, ">:utf8", $a_file or die $!;
binmode(F); # we write a "\n" and then tell() - avoid CRLF issues.
binmode(F,":utf8"); # turn UTF-8-ness back on
print F $a;
my $y;
{ my $x = tell(F);
{ use bytes; $y = length($a);}
- ok( $x == $y );
+ cmp_ok( $x, '==', $y );
}
{ # Check byte length of $b
use bytes; my $y = length($b);
- ok( $y == 1 );
+ cmp_ok( $y, '==', 1 );
}
print F $b,"\n"; # Don't upgrades $b
{ # Check byte length of $b
use bytes; my $y = length($b);
- ok( $y == 1 );
+ cmp_ok( $y, '==', 1 );
}
{
my $x = tell(F);
{ use bytes; if (ord('A')==193){$y += 2;}else{$y += 3;}} # EBCDIC ASCII
- ok( $x == $y );
+ cmp_ok( $x, '==', $y );
}
close F;
- open F, "a" or die $!; # Not UTF
+ open F, $a_file or die $!; # Not UTF
binmode(F, ":bytes");
$x = <F>;
chomp($x);
$chr = v196.172.194.130;
if (ord('A') == 193) { $chr = v141.83.130; } # EBCDIC
- ok( $x eq $chr, sprintf('(%vd)', $x) );
+ is( $x, $chr, sprintf('(%vd)', $x) );
- open F, "<:utf8", "a" or die $!;
+ open F, "<:utf8", $a_file or die $!;
$x = <F>;
chomp($x);
close F;
- ok( $x eq chr(300).chr(130), sprintf('(%vd)', $x) );
+ is( $x, chr(300).chr(130), sprintf('(%vd)', $x) );
- open F, ">", "a" or die $!;
- if (${^OPEN} =~ /:utf8/) {
- binmode(F, ":bytes:");
- }
+ open F, ">", $a_file or die $!;
+ binmode(F, ":bytes:");
# Now let's make it suffer.
my $w;
use warnings 'utf8';
local $SIG{__WARN__} = sub { $w = $_[0] };
print F $a;
- ok( !($@ || $w !~ /Wide character in print/i) );
+ ok( (!$@));
+ like($w, qr/Wide character in print/i );
}
}
# Hm. Time to get more evil.
-open F, ">:utf8", "a" or die $!;
+open F, ">:utf8", $a_file or die $!;
print F $a;
binmode(F, ":bytes");
print F chr(130)."\n";
close F;
-open F, "<", "a" or die $!;
+open F, "<", $a_file or die $!;
binmode(F, ":bytes");
$x = <F>; chomp $x;
$chr = v196.172.130;
if (ord('A') == 193) { $chr = v141.83.130; } # EBCDIC
-ok( $x eq $chr );
+is( $x, $chr );
# Right.
-open F, ">:utf8", "a" or die $!;
+open F, ">:utf8", $a_file or die $!;
print F $a;
close F;
-open F, ">>", "a" or die $!;
+open F, ">>", $a_file or die $!;
+binmode(F, ":bytes");
print F chr(130)."\n";
close F;
-open F, "<", "a" or die $!;
+open F, "<", $a_file or die $!;
+binmode(F, ":bytes");
$x = <F>; chomp $x;
-ok( $x eq $chr );
+SKIP: {
+ skip("Defaulting to UTF-8 output means that we can't generate a mangled file")
+ if $UTF8_OUTPUT;
+ is( $x, $chr );
+}
# Now we have a deformed file.
SKIP: {
if (ord('A') == 193) {
- skip( "EBCDIC doesn't complain" );
+ skip("EBCDIC doesn't complain", 2);
} else {
- open F, "<:utf8", "a" or die $!;
+ my @warnings;
+ open F, "<:utf8", $a_file or die $!;
$x = <F>; chomp $x;
- local $SIG{__WARN__} = sub { ok( 1 ) };
+ local $SIG{__WARN__} = sub { push @warnings, $_[0]; };
eval { sprintf "%vd\n", $x };
+ is (scalar @warnings, 1);
+ like ($warnings[0], qr/Malformed UTF-8 character \(unexpected continuation byte 0x82, with no preceding start byte/);
}
}
close F;
-unlink('a');
+unlink($a_file);
-open F, ">:utf8", "a";
+open F, ">:utf8", $a_file;
@a = map { chr(1 << ($_ << 2)) } 0..5; # 0x1, 0x10, .., 0x100000
unshift @a, chr(0); # ... and a null byte in front just for fun
print F @a;
my $c;
# read() should work on characters, not bytes
-open F, "<:utf8", "a";
+open F, "<:utf8", $a_file;
$a = 0;
+my $failed;
for (@a) {
unless (($c = read(F, $b, 1) == 1) &&
length($b) == 1 &&
print '# tell(F) == ', tell(F), "\n";
print '# $a == ', $a, "\n";
print '# $c == ', $c, "\n";
- print "not ";
+ $failed++;
last;
}
}
close F;
-ok( 1 );
+is($failed, undef);
{
# Check that warnings are on on I/O, and that they can be muffled.
local $SIG{__WARN__} = sub { $@ = shift };
undef $@;
- open F, ">a";
+ open F, ">$a_file";
binmode(F, ":bytes");
print F chr(0x100);
close(F);
like( $@, 'Wide character in print' );
undef $@;
- open F, ">:utf8", "a";
+ open F, ">:utf8", $a_file;
print F chr(0x100);
close(F);
- isnt( defined $@ );
+ isnt( defined $@, !0 );
undef $@;
- open F, ">a";
+ open F, ">$a_file";
binmode(F, ":utf8");
print F chr(0x100);
close(F);
- isnt( defined $@ );
+ isnt( defined $@, !0 );
no warnings 'utf8';
undef $@;
- open F, ">a";
+ open F, ">$a_file";
print F chr(0x100);
close(F);
- isnt( defined $@ );
+ isnt( defined $@, !0 );
use warnings 'utf8';
undef $@;
- open F, ">a";
+ open F, ">$a_file";
binmode(F, ":bytes");
print F chr(0x100);
close(F);
}
{
- open F, ">:bytes","a"; print F "\xde"; close F;
+ open F, ">:bytes",$a_file; print F "\xde"; close F;
- open F, "<:bytes", "a";
+ open F, "<:bytes", $a_file;
my $b = chr 0x100;
$b .= <F>;
- ok( $b eq chr(0x100).chr(0xde), "21395 '.= <>' utf8 vs. bytes" );
+ is( $b, chr(0x100).chr(0xde), "21395 '.= <>' utf8 vs. bytes" );
close F;
}
{
- open F, ">:utf8","a"; print F chr 0x100; close F;
+ open F, ">:utf8",$a_file; print F chr 0x100; close F;
- open F, "<:utf8", "a";
+ open F, "<:utf8", $a_file;
my $b = "\xde";
$b .= <F>;
- ok( $b eq chr(0xde).chr(0x100), "21395 '.= <>' bytes vs. utf8" );
+ is( $b, chr(0xde).chr(0x100), "21395 '.= <>' bytes vs. utf8" );
close F;
}
for my $u (@a) {
for my $v (@a) {
# print "# @$u - @$v\n";
- open F, ">a";
+ open F, ">$a_file";
binmode(F, ":" . $u->[1]);
print F chr($u->[0]);
close F;
- open F, "<a";
+ open F, "<$a_file";
binmode(F, ":" . $u->[1]);
my $s = chr($v->[0]);
utf8::upgrade($s) if $v->[1] eq "utf8";
$s .= <F>;
- ok( $s eq chr($v->[0]) . chr($u->[0]), 'rcatline utf8' );
+ is( $s, chr($v->[0]) . chr($u->[0]), 'rcatline utf8' );
close F;
$t++;
}
{
# [perl #23428] Somethings rotten in unicode semantics
- open F, ">a";
+ open F, ">$a_file";
binmode F, ":utf8";
syswrite(F, $a = chr(0x100));
close F;
# <FH> on a :utf8 stream should complain immediately with -w
# if it finds bad UTF-8 (:encoding(utf8) works this way)
use warnings 'utf8';
+ undef $@;
local $SIG{__WARN__} = sub { $@ = shift };
- open F, ">a";
+ open F, ">$a_file";
binmode F;
- print F "foo", chr(0xE4), "\n";
+ my ($chrE4, $chrF6) = (chr(0xE4), chr(0xF6));
+ if (ord('A') == 193) # EBCDIC
+ { ($chrE4, $chrF6) = (chr(0x43), chr(0xEC)); }
+ print F "foo", $chrE4, "\n";
+ print F "foo", $chrF6, "\n";
close F;
- open F, "<:utf8", "a";
+ open F, "<:utf8", $a_file;
+ undef $@;
my $line = <F>;
- like( $@, qr/utf8 "\\xE4" does not map to Unicode/ );
+ my ($chrE4, $chrF6) = ("E4", "F6");
+ if (ord('A') == 193) { ($chrE4, $chrF6) = ("43", "EC"); } # EBCDIC
+ like( $@, qr/utf8 "\\x$chrE4" does not map to Unicode .+ <F> line 1/,
+ "<:utf8 readline must warn about bad utf8");
+ undef $@;
+ $line .= <F>;
+ like( $@, qr/utf8 "\\x$chrF6" does not map to Unicode .+ <F> line 2/,
+ "<:utf8 rcatline must warn about bad utf8");
close F;
}
-
-END {
- 1 while unlink "a";
- 1 while unlink "b";
-}
projects / p5sagit/p5-mst-13.2.git / blobdiff