#!./perl
-print "1..44\n";
+BEGIN {
+ chdir 't' if -d 't';
+ @INC = '../lib';
+ require './test.pl';
+}
+
+plan tests => 251;
$FS = ':';
($a,$b,$c) = split($FS,$_);
-if (join(';',$a,$b,$c) eq 'a;b;c') {print "ok 1\n";} else {print "not ok 1\n";}
+is(join(';',$a,$b,$c), 'a;b;c');
@ary = split(/:b:/);
-if (join("$_",@ary) eq 'aa:b:cc') {print "ok 2\n";} else {print "not ok 2\n";}
+$cnt = split(/:b:/);
+is(join("$_",@ary), 'aa:b:cc');
+is($cnt, scalar(@ary));
$_ = "abc\n";
my @xyz = (@ary = split(//));
-if (join(".",@ary) eq "a.b.c.\n") {print "ok 3\n";} else {print "not ok 3\n";}
+$cnt = split(//);
+is(join(".",@ary), "a.b.c.\n");
+is($cnt, scalar(@ary));
$_ = "a:b:c::::";
@ary = split(/:/);
-if (join(".",@ary) eq "a.b.c") {print "ok 4\n";} else {print "not ok 4\n";}
+$cnt = split(/:/);
+is(join(".",@ary), "a.b.c");
+is($cnt, scalar(@ary));
$_ = join(':',split(' '," a b\tc \t d "));
-if ($_ eq 'a:b:c:d') {print "ok 5\n";} else {print "not ok 5 #$_#\n";}
+is($_, 'a:b:c:d');
+@ary = split(' '," a b\tc \t d ");
+$cnt = split(' '," a b\tc \t d ");
+is($cnt, scalar(@ary));
$_ = join(':',split(/ */,"foo bar bie\tdoll"));
-if ($_ eq "f:o:o:b:a:r:b:i:e:\t:d:o:l:l")
- {print "ok 6\n";} else {print "not ok 6\n";}
+is($_ , "f:o:o:b:a:r:b:i:e:\t:d:o:l:l");
+@ary = split(/ */,"foo bar bie\tdoll");
+$cnt = split(/ */,"foo bar bie\tdoll");
+is($cnt, scalar(@ary));
$_ = join(':', 'foo', split(/ /,'a b c'), 'bar');
-if ($_ eq "foo:a:b::c:bar") {print "ok 7\n";} else {print "not ok 7 $_\n";}
+is($_, "foo:a:b::c:bar");
+@ary = split(/ /,'a b c');
+$cnt = split(/ /,'a b c');
+is($cnt, scalar(@ary));
# Can we say how many fields to split to?
$_ = join(':', split(' ','1 2 3 4 5 6', 3));
-print $_ eq '1:2:3 4 5 6' ? "ok 8\n" : "not ok 8 $_\n";
+is($_, '1:2:3 4 5 6');
+@ary = split(' ','1 2 3 4 5 6', 3);
+$cnt = split(' ','1 2 3 4 5 6', 3);
+is($cnt, scalar(@ary));
# Can we do it as a variable?
$x = 4;
$_ = join(':', split(' ','1 2 3 4 5 6', $x));
-print $_ eq '1:2:3:4 5 6' ? "ok 9\n" : "not ok 9 $_\n";
+is($_, '1:2:3:4 5 6');
+@ary = split(' ','1 2 3 4 5 6', $x);
+$cnt = split(' ','1 2 3 4 5 6', $x);
+is($cnt, scalar(@ary));
# Does the 999 suppress null field chopping?
$_ = join(':', split(/:/,'1:2:3:4:5:6:::', 999));
-print $_ eq '1:2:3:4:5:6:::' ? "ok 10\n" : "not ok 10 $_\n";
+is($_ , '1:2:3:4:5:6:::');
+@ary = split(/:/,'1:2:3:4:5:6:::', 999);
+$cnt = split(/:/,'1:2:3:4:5:6:::', 999);
+is($cnt, scalar(@ary));
# Does assignment to a list imply split to one more field than that?
-if ($^O eq 'MSWin32') { $foo = `.\\perl -D1024 -e "(\$a,\$b) = split;" 2>&1` }
-elsif ($^O eq 'VMS') { $foo = `./perl "-D1024" -e "(\$a,\$b) = split;" 2>&1` }
-else { $foo = `./perl -D1024 -e '(\$a,\$b) = split;' 2>&1` }
-print $foo =~ /DEBUGGING/ || $foo =~ /SV = (VOID|IV\(3\))/ ? "ok 11\n" : "not ok 11\n";
+$foo = runperl( switches => ['-Dt'], stderr => 1, prog => '($a,$b)=split;' );
+ok($foo =~ /DEBUGGING/ || $foo =~ /const\n?\Q(IV(3))\E/);
# Can we say how many fields to split to when assigning to a list?
($a,$b) = split(' ','1 2 3 4 5 6', 2);
$_ = join(':',$a,$b);
-print $_ eq '1:2 3 4 5 6' ? "ok 12\n" : "not ok 12 $_\n";
+is($_, '1:2 3 4 5 6');
# do subpatterns generate additional fields (without trailing nulls)?
$_ = join '|', split(/,|(-)/, "1-10,20,,,");
-print $_ eq "1|-|10||20" ? "ok 13\n" : "not ok 13\n";
+is($_, "1|-|10||20");
+@ary = split(/,|(-)/, "1-10,20,,,");
+$cnt = split(/,|(-)/, "1-10,20,,,");
+is($cnt, scalar(@ary));
# do subpatterns generate additional fields (with a limit)?
$_ = join '|', split(/,|(-)/, "1-10,20,,,", 10);
-print $_ eq "1|-|10||20||||||" ? "ok 14\n" : "not ok 14\n";
+is($_, "1|-|10||20||||||");
+@ary = split(/,|(-)/, "1-10,20,,,", 10);
+$cnt = split(/,|(-)/, "1-10,20,,,", 10);
+is($cnt, scalar(@ary));
# is the 'two undefs' bug fixed?
(undef, $a, undef, $b) = qw(1 2 3 4);
-print "$a|$b" eq "2|4" ? "ok 15\n" : "not ok 15\n";
+is("$a|$b", "2|4");
# .. even for locals?
{
local(undef, $a, undef, $b) = qw(1 2 3 4);
- print "$a|$b" eq "2|4" ? "ok 16\n" : "not ok 16\n";
+ is("$a|$b", "2|4");
}
# check splitting of null string
$_ = join('|', split(/x/, '',-1), 'Z');
-print $_ eq "Z" ? "ok 17\n" : "#$_\nnot ok 17\n";
+is($_, "Z");
+@ary = split(/x/, '',-1);
+$cnt = split(/x/, '',-1);
+is($cnt, scalar(@ary));
$_ = join('|', split(/x/, '', 1), 'Z');
-print $_ eq "Z" ? "ok 18\n" : "#$_\nnot ok 18\n";
+is($_, "Z");
+@ary = split(/x/, '', 1);
+$cnt = split(/x/, '', 1);
+is($cnt, scalar(@ary));
$_ = join('|', split(/(p+)/,'',-1), 'Z');
-print $_ eq "Z" ? "ok 19\n" : "#$_\nnot ok 19\n";
+is($_, "Z");
+@ary = split(/(p+)/,'',-1);
+$cnt = split(/(p+)/,'',-1);
+is($cnt, scalar(@ary));
$_ = join('|', split(/.?/, '',-1), 'Z');
-print $_ eq "Z" ? "ok 20\n" : "#$_\nnot ok 20\n";
+is($_, "Z");
+@ary = split(/.?/, '',-1);
+$cnt = split(/.?/, '',-1);
+is($cnt, scalar(@ary));
# Are /^/m patterns scanned?
$_ = join '|', split(/^a/m, "a b a\na d a", 20);
-print $_ eq "| b a\n| d a" ? "ok 21\n" : "not ok 21\n# `$_'\n";
+is($_, "| b a\n| d a");
+@ary = split(/^a/m, "a b a\na d a", 20);
+$cnt = split(/^a/m, "a b a\na d a", 20);
+is($cnt, scalar(@ary));
# Are /$/m patterns scanned?
$_ = join '|', split(/a$/m, "a b a\na d a", 20);
-print $_ eq "a b |\na d |" ? "ok 22\n" : "not ok 22\n# `$_'\n";
+is($_, "a b |\na d |");
+@ary = split(/a$/m, "a b a\na d a", 20);
+$cnt = split(/a$/m, "a b a\na d a", 20);
+is($cnt, scalar(@ary));
# Are /^/m patterns scanned?
$_ = join '|', split(/^aa/m, "aa b aa\naa d aa", 20);
-print $_ eq "| b aa\n| d aa" ? "ok 23\n" : "not ok 23\n# `$_'\n";
+is($_, "| b aa\n| d aa");
+@ary = split(/^aa/m, "aa b aa\naa d aa", 20);
+$cnt = split(/^aa/m, "aa b aa\naa d aa", 20);
+is($cnt, scalar(@ary));
# Are /$/m patterns scanned?
$_ = join '|', split(/aa$/m, "aa b aa\naa d aa", 20);
-print $_ eq "aa b |\naa d |" ? "ok 24\n" : "not ok 24\n# `$_'\n";
+is($_, "aa b |\naa d |");
+@ary = split(/aa$/m, "aa b aa\naa d aa", 20);
+$cnt = split(/aa$/m, "aa b aa\naa d aa", 20);
+is($cnt, scalar(@ary));
# Greedyness:
$_ = "a : b :c: d";
@ary = split(/\s*:\s*/);
-if (($res = join(".",@ary)) eq "a.b.c.d") {print "ok 25\n";} else {print "not ok 25\n# res=`$res' != `a.b.c.d'\n";}
+$cnt = split(/\s*:\s*/);
+is(($res = join(".",@ary)), "a.b.c.d", $res);
+is($cnt, scalar(@ary));
# use of match result as pattern (!)
-'p:q:r:s' eq join ':', split('abc' =~ /b/, 'p1q1r1s') or print "not ";
-print "ok 26\n";
+is('p:q:r:s', join ':', split('abc' =~ /b/, 'p1q1r1s'));
+@ary = split('abc' =~ /b/, 'p1q1r1s');
+$cnt = split('abc' =~ /b/, 'p1q1r1s');
+is($cnt, scalar(@ary));
# /^/ treated as /^/m
$_ = join ':', split /^/, "ab\ncd\nef\n";
-print "not " if $_ ne "ab\n:cd\n:ef\n";
-print "ok 27\n";
+is($_, "ab\n:cd\n:ef\n");
# see if @a = @b = split(...) optimization works
@list1 = @list2 = split ('p',"a p b c p");
-print "not " if @list1 != @list2 or "@list1" ne "@list2"
- or @list1 != 2 or "@list1" ne "a b c ";
-print "ok 28\n";
+ok(@list1 == @list2 &&
+ "@list1" eq "@list2" &&
+ @list1 == 2 &&
+ "@list1" eq "a b c ");
# zero-width assertion
$_ = join ':', split /(?=\w)/, "rm b";
-print "not" if $_ ne "r:m :b";
-print "ok 29\n";
+is($_, "r:m :b");
+@ary = split /(?=\w)/, "rm b";
+$cnt = split /(?=\w)/, "rm b";
+is($cnt, scalar(@ary));
# unicode splittage
@ary = map {ord} split //, v1.20.300.4000.50000.4000.300.20.1;
-print "not " unless "@ary" eq "1 20 300 4000 50000 4000 300 20 1";
-print "ok 30\n";
+$cnt = split //, v1.20.300.4000.50000.4000.300.20.1;
+is("@ary", "1 20 300 4000 50000 4000 300 20 1");
+is($cnt, scalar(@ary));
@ary = split(/\x{FE}/, "\x{FF}\x{FE}\x{FD}"); # bug id 20010105.016
-print "not " unless @ary == 2 &&
- $ary[0] eq "\xFF" && $ary[1] eq "\xFD" &&
- $ary[0] eq "\x{FF}" && $ary[1] eq "\x{FD}";
-print "ok 31\n";
+$cnt = split(/\x{FE}/, "\x{FF}\x{FE}\x{FD}"); # bug id 20010105.016
+ok(@ary == 2 &&
+ $ary[0] eq "\xFF" && $ary[1] eq "\xFD" &&
+ $ary[0] eq "\x{FF}" && $ary[1] eq "\x{FD}");
+is($cnt, scalar(@ary));
@ary = split(/(\x{FE}\xFE)/, "\xFF\x{FF}\xFE\x{FE}\xFD\x{FD}"); # variant of 31
-print "not " unless @ary == 3 &&
- $ary[0] eq "\xFF\xFF" &&
- $ary[0] eq "\x{FF}\xFF" &&
- $ary[0] eq "\x{FF}\x{FF}" &&
- $ary[1] eq "\xFE\xFE" &&
- $ary[1] eq "\x{FE}\xFE" &&
- $ary[1] eq "\x{FE}\x{FE}" &&
- $ary[2] eq "\xFD\xFD" &&
- $ary[2] eq "\x{FD}\xFD" &&
- $ary[2] eq "\x{FD}\x{FD}";
-print "ok 32\n";
-
+$cnt = split(/(\x{FE}\xFE)/, "\xFF\x{FF}\xFE\x{FE}\xFD\x{FD}"); # variant of 31
+ok(@ary == 3 &&
+ $ary[0] eq "\xFF\xFF" &&
+ $ary[0] eq "\x{FF}\xFF" &&
+ $ary[0] eq "\x{FF}\x{FF}" &&
+ $ary[1] eq "\xFE\xFE" &&
+ $ary[1] eq "\x{FE}\xFE" &&
+ $ary[1] eq "\x{FE}\x{FE}" &&
+ $ary[2] eq "\xFD\xFD" &&
+ $ary[2] eq "\x{FD}\xFD" &&
+ $ary[2] eq "\x{FD}\x{FD}");
+is($cnt, scalar(@ary));
{
my @a = map ord, split(//, join("", map chr, (1234, 123, 2345)));
- print "not " unless "@a" eq "1234 123 2345";
- print "ok 33\n";
+ my $c = split(//, join("", map chr, (1234, 123, 2345)));
+ is("@a", "1234 123 2345");
+ is($c, scalar(@a));
}
{
- my $x = chr(123);
- my @a = map ord, split(/$x/, join("", map chr, (1234, 123, 2345)));
- print "not " unless "@a" eq "1234 2345";
- print "ok 34\n";
+ my $x = 'A';
+ my @a = map ord, split(/$x/, join("", map chr, (1234, ord($x), 2345)));
+ my $c = split(/$x/, join("", map chr, (1234, ord($x), 2345)));
+ is("@a", "1234 2345");
+ is($c, scalar(@a));
}
{
my $sushi = "\x{b36c}\x{5a8c}\x{ff5b}\x{5079}\x{505b}";
my @charlist = split //, $sushi;
+ my $charnum = split //, $sushi;
+ is($charnum, scalar(@charlist));
my $r = '';
foreach my $ch (@charlist) {
$r = $r . " " . sprintf "U+%04X", ord($ch);
}
- print "not " unless $r eq " U+B36C U+5A8C U+FF5B U+5079 U+505B";
- print "ok 35\n";
+ is($r, " U+B36C U+5A8C U+FF5B U+5079 U+505B");
}
{
- # bug id 20000426.003
-
my $s = "\x20\x40\x{80}\x{100}\x{80}\x40\x20";
- my ($a, $b, $c) = split(/\x40/, $s);
- print "not "
- unless $a eq "\x20" && $b eq "\x{80}\x{100}\x{80}" && $c eq $a;
- print "ok 36\n";
+ SKIP: {
+ if (ord('A') == 193) {
+ skip("EBCDIC", 1);
+ } else {
+ # bug id 20000426.003
+
+ my ($a, $b, $c) = split(/\x40/, $s);
+ ok($a eq "\x20" && $b eq "\x{80}\x{100}\x{80}" && $c eq $a);
+ }
+ }
my ($a, $b) = split(/\x{100}/, $s);
- print "not " unless $a eq "\x20\x40\x{80}" && $b eq "\x{80}\x40\x20";
- print "ok 37\n";
+ ok($a eq "\x20\x40\x{80}" && $b eq "\x{80}\x40\x20");
my ($a, $b) = split(/\x{80}\x{100}\x{80}/, $s);
- print "not " unless $a eq "\x20\x40" && $b eq "\x40\x20";
- print "ok 38\n";
-
- my ($a, $b) = split(/\x40\x{80}/, $s);
- print "not " unless $a eq "\x20" && $b eq "\x{100}\x{80}\x40\x20";
- print "ok 39\n";
+ ok($a eq "\x20\x40" && $b eq "\x40\x20");
+
+ SKIP: {
+ if (ord('A') == 193) {
+ skip("EBCDIC", 1);
+ } else {
+ my ($a, $b) = split(/\x40\x{80}/, $s);
+ ok($a eq "\x20" && $b eq "\x{100}\x{80}\x40\x20");
+ }
+ }
my ($a, $b, $c) = split(/[\x40\x{80}]+/, $s);
- print "not " unless $a eq "\x20" && $b eq "\x{100}" && $c eq "\x20";
- print "ok 40\n";
+ ok($a eq "\x20" && $b eq "\x{100}" && $c eq "\x20");
}
{
my $a = "ABC\x{263A}";
my @b = split( //, $a );
+ my $c = split( //, $a );
+ is($c, scalar(@b));
- print "not " unless @b == 4;
- print "ok 41\n";
+ is(scalar @b, 4);
- print "not " unless length($b[3]) == 1 && $b[3] eq "\x{263A}";
- print "ok 42\n";
+ ok(length($b[3]) == 1 && $b[3] eq "\x{263A}");
$a =~ s/^A/Z/;
- print "not " unless length($a) == 4 && $a eq "ZBC\x{263A}";
- print "ok 43\n";
+ ok(length($a) == 4 && $a eq "ZBC\x{263A}");
}
{
my @a = split(/\xFE/, "\xFF\xFE\xFD");
+ my $b = split(/\xFE/, "\xFF\xFE\xFD");
+
+ ok(@a == 2 && $a[0] eq "\xFF" && $a[1] eq "\xFD");
+ is($b, scalar(@a));
+}
+
+{
+ # check that PMf_WHITE is cleared after \s+ is used
+ # reported in <20010627113312.RWGY6087.viemta06@localhost>
+ my $r;
+ foreach my $pat ( qr/\s+/, qr/ll/ ) {
+ $r = join ':' => split($pat, "hello cruel world");
+ }
+ is($r, "he:o cruel world");
+}
- print "not " unless @a == 2 && $a[0] eq "\xFF" && $a[1] eq "\xFD";
- print "ok 44\n";
+
+{
+ # split /(A)|B/, "1B2" should return (1, undef, 2)
+ my @x = split /(A)|B/, "1B2";
+ my $y = split /(A)|B/, "1B2";
+ is($y, scalar(@x));
+ ok($x[0] eq '1' and (not defined $x[1]) and $x[2] eq '2');
}
+{
+ # [perl #17064]
+ my $warn;
+ local $SIG{__WARN__} = sub { $warn = join '', @_; chomp $warn };
+ my $char = "\x{10f1ff}";
+ my @a = split /\r?\n/, "$char\n";
+ my $b = split /\r?\n/, "$char\n";
+ is($b, scalar(@a));
+ ok(@a == 1 && $a[0] eq $char && !defined($warn));
+}
+
+{
+ # [perl #18195]
+ for my $u (0, 1) {
+ for my $a (0, 1) {
+ $_ = 'readin,database,readout';
+ utf8::upgrade $_ if $u;
+ /(.+)/;
+ my @d = split /[,]/,$1;
+ my $e = split /[,]/,$1;
+ is($e, scalar(@d));
+ is(join (':',@d), 'readin:database:readout', "[perl #18195]");
+ }
+ }
+}
+
+{
+ $p="a,b";
+ utf8::upgrade $p;
+ eval { @a=split(/[, ]+/,$p) };
+ eval { $b=split(/[, ]+/,$p) };
+ is($b, scalar(@a));
+ is ("$@-@a-", '-a b-', '#20912 - split() to array with /[]+/ and utf8');
+}
+
+{
+ is (\@a, \@{"a"}, '@a must be global for following test');
+ $p="";
+ $n = @a = split /,/,$p;
+ is ($n, 0, '#21765 - pmreplroot hack used to return undef for 0 iters');
+}
+
+{
+ # [perl #28938]
+ # assigning off the end of the array after a split could leave garbage
+ # in the inner elements
+
+ my $x;
+ @a = split /,/, ',,,,,';
+ $a[3]=1;
+ $x = \$a[2];
+ is (ref $x, 'SCALAR', '#28938 - garbage after extend');
+}
+{
+ # check the special casing of split /\s/ and unicode
+ use charnames qw(:full);
+ # below test data is extracted from
+ # PropList-5.0.0.txt
+ # Date: 2006-06-07, 23:22:52 GMT [MD]
+ #
+ # Unicode Character Database
+ # Copyright (c) 1991-2006 Unicode, Inc.
+ # For terms of use, see http://www.unicode.org/terms_of_use.html
+ # For documentation, see UCD.html
+ my @spaces=(
+ ord("\t"), # Cc <control-0009>
+ ord("\n"), # Cc <control-000A>
+ # not PerlSpace # Cc <control-000B>
+ ord("\f"), # Cc <control-000C>
+ ord("\r"), # Cc <control-000D>
+ ord(" "), # Zs SPACE
+ ord("\N{NEL}"), # Cc <control-0085>
+ ord("\N{NO-BREAK SPACE}"),
+ # Zs NO-BREAK SPACE
+ 0x1680, # Zs OGHAM SPACE MARK
+ 0x180E, # Zs MONGOLIAN VOWEL SEPARATOR
+ 0x2000..0x200A, # Zs [11] EN QUAD..HAIR SPACE
+ 0x2028, # Zl LINE SEPARATOR
+ 0x2029, # Zp PARAGRAPH SEPARATOR
+ 0x202F, # Zs NARROW NO-BREAK SPACE
+ 0x205F, # Zs MEDIUM MATHEMATICAL SPACE
+ 0x3000 # Zs IDEOGRAPHIC SPACE
+ );
+ #diag "Have @{[0+@spaces]} to test\n";
+ foreach my $cp (@spaces) {
+ my $msg = sprintf "Space: U+%04x", $cp;
+ my $space = chr($cp);
+ my $str="A:$space:B\x{FFFD}";
+ chop $str;
+
+ my @res=split(/\s+/,$str);
+ my $cnt=split(/\s+/,$str);
+ ok(@res == 2 && join('-',@res) eq "A:-:B", "$msg - /\\s+/");
+ is($cnt, scalar(@res), "$msg - /\\s+/ (count)");
+
+ my $s2 = "$space$space:A:$space$space:B\x{FFFD}";
+ chop $s2;
+
+ my @r2 = split(' ',$s2);
+ my $c2 = split(' ',$s2);
+ ok(@r2 == 2 && join('-', @r2) eq ":A:-:B", "$msg - ' '");
+ is($c2, scalar(@r2), "$msg - ' ' (count)");
+
+ my @r3 = split(/\s+/, $s2);
+ my $c3 = split(/\s+/, $s2);
+ ok(@r3 == 3 && join('-', @r3) eq "-:A:-:B", "$msg - /\\s+/ No.2");
+ is($c3, scalar(@r3), "$msg - /\\s+/ No.2 (count)");
+ }
+}
+
+{
+ my $src = "ABC \0 FOO \0 XYZ";
+ my @s = split(" \0 ", $src);
+ my @r = split(/ \0 /, $src);
+ my $cs = split(" \0 ", $src);
+ my $cr = split(/ \0 /, $src);
+ is(scalar(@s), 3);
+ is($cs, 3);
+ is($cr, 3);
+ is($s[0], "ABC");
+ is($s[1], "FOO");
+ is($s[2]," XYZ");
+ is(join(':',@s), join(':',@r));
+}
+
+{
+ use constant BANG => {};
+ () = split m/,/, "", BANG;
+ ok(1);
+}
+
+{
+ # Bug #69875
+ # 'Hybrid' scalar-and-array context
+ scalar(our @PATH = split /::/, "Font::GlyphNames");
+ # 'my' doesn't trigger the bug
+ is "@PATH", "Font GlyphNames", "hybrid scalar-and-array context";
+}
https://perl5.git.perl.org / perl5.git / blobdiff