1#!./perl 2 3BEGIN { 4 chdir 't' if -d 't'; 5 @INC = '../lib'; 6 require './test.pl'; 7} 8 9plan tests => 55; 10 11$FS = ':'; 12 13$_ = 'a:b:c'; 14 15($a,$b,$c) = split($FS,$_); 16 17is(join(';',$a,$b,$c), 'a;b;c'); 18 19@ary = split(/:b:/); 20is(join("$_",@ary), 'aa:b:cc'); 21 22$_ = "abc\n"; 23my @xyz = (@ary = split(//)); 24is(join(".",@ary), "a.b.c.\n"); 25 26$_ = "a:b:c::::"; 27@ary = split(/:/); 28is(join(".",@ary), "a.b.c"); 29 30$_ = join(':',split(' '," a b\tc \t d ")); 31is($_, 'a:b:c:d'); 32 33$_ = join(':',split(/ */,"foo bar bie\tdoll")); 34is($_ , "f:o:o:b:a:r:b:i:e:\t:d:o:l:l"); 35 36$_ = join(':', 'foo', split(/ /,'a b c'), 'bar'); 37is($_, "foo:a:b::c:bar"); 38 39# Can we say how many fields to split to? 40$_ = join(':', split(' ','1 2 3 4 5 6', 3)); 41is($_, '1:2:3 4 5 6'); 42 43# Can we do it as a variable? 44$x = 4; 45$_ = join(':', split(' ','1 2 3 4 5 6', $x)); 46is($_, '1:2:3:4 5 6'); 47 48# Does the 999 suppress null field chopping? 49$_ = join(':', split(/:/,'1:2:3:4:5:6:::', 999)); 50is($_ , '1:2:3:4:5:6:::'); 51 52# Does assignment to a list imply split to one more field than that? 53$foo = runperl( switches => ['-Dt'], stderr => 1, prog => '($a,$b)=split;' ); 54ok($foo =~ /DEBUGGING/ || $foo =~ /const\n?\Q(IV(3))\E/); 55 56# Can we say how many fields to split to when assigning to a list? 57($a,$b) = split(' ','1 2 3 4 5 6', 2); 58$_ = join(':',$a,$b); 59is($_, '1:2 3 4 5 6'); 60 61# do subpatterns generate additional fields (without trailing nulls)? 62$_ = join '|', split(/,|(-)/, "1-10,20,,,"); 63is($_, "1|-|10||20"); 64 65# do subpatterns generate additional fields (with a limit)? 66$_ = join '|', split(/,|(-)/, "1-10,20,,,", 10); 67is($_, "1|-|10||20||||||"); 68 69# is the 'two undefs' bug fixed? 70(undef, $a, undef, $b) = qw(1 2 3 4); 71is("$a|$b", "2|4"); 72 73# .. even for locals? 74{ 75 local(undef, $a, undef, $b) = qw(1 2 3 4); 76 is("$a|$b", "2|4"); 77} 78 79# check splitting of null string 80$_ = join('|', split(/x/, '',-1), 'Z'); 81is($_, "Z"); 82 83$_ = join('|', split(/x/, '', 1), 'Z'); 84is($_, "Z"); 85 86$_ = join('|', split(/(p+)/,'',-1), 'Z'); 87is($_, "Z"); 88 89$_ = join('|', split(/.?/, '',-1), 'Z'); 90is($_, "Z"); 91 92 93# Are /^/m patterns scanned? 94$_ = join '|', split(/^a/m, "a b a\na d a", 20); 95is($_, "| b a\n| d a"); 96 97# Are /$/m patterns scanned? 98$_ = join '|', split(/a$/m, "a b a\na d a", 20); 99is($_, "a b |\na d |"); 100 101# Are /^/m patterns scanned? 102$_ = join '|', split(/^aa/m, "aa b aa\naa d aa", 20); 103is($_, "| b aa\n| d aa"); 104 105# Are /$/m patterns scanned? 106$_ = join '|', split(/aa$/m, "aa b aa\naa d aa", 20); 107is($_, "aa b |\naa d |"); 108 109# Greedyness: 110$_ = "a : b :c: d"; 111@ary = split(/\s*:\s*/); 112is(($res = join(".",@ary)), "a.b.c.d", $res); 113 114# use of match result as pattern (!) 115is('p:q:r:s', join ':', split('abc' =~ /b/, 'p1q1r1s')); 116 117# /^/ treated as /^/m 118$_ = join ':', split /^/, "ab\ncd\nef\n"; 119is($_, "ab\n:cd\n:ef\n"); 120 121# see if @a = @b = split(...) optimization works 122@list1 = @list2 = split ('p',"a p b c p"); 123ok(@list1 == @list2 && 124 "@list1" eq "@list2" && 125 @list1 == 2 && 126 "@list1" eq "a b c "); 127 128# zero-width assertion 129$_ = join ':', split /(?=\w)/, "rm b"; 130is($_, "r:m :b"); 131 132# unicode splittage 133 134@ary = map {ord} split //, v1.20.300.4000.50000.4000.300.20.1; 135is("@ary", "1 20 300 4000 50000 4000 300 20 1"); 136 137@ary = split(/\x{FE}/, "\x{FF}\x{FE}\x{FD}"); # bug id 20010105.016 138ok(@ary == 2 && 139 $ary[0] eq "\xFF" && $ary[1] eq "\xFD" && 140 $ary[0] eq "\x{FF}" && $ary[1] eq "\x{FD}"); 141 142@ary = split(/(\x{FE}\xFE)/, "\xFF\x{FF}\xFE\x{FE}\xFD\x{FD}"); # variant of 31 143ok(@ary == 3 && 144 $ary[0] eq "\xFF\xFF" && 145 $ary[0] eq "\x{FF}\xFF" && 146 $ary[0] eq "\x{FF}\x{FF}" && 147 $ary[1] eq "\xFE\xFE" && 148 $ary[1] eq "\x{FE}\xFE" && 149 $ary[1] eq "\x{FE}\x{FE}" && 150 $ary[2] eq "\xFD\xFD" && 151 $ary[2] eq "\x{FD}\xFD" && 152 $ary[2] eq "\x{FD}\x{FD}"); 153 154{ 155 my @a = map ord, split(//, join("", map chr, (1234, 123, 2345))); 156 is("@a", "1234 123 2345"); 157} 158 159{ 160 my $x = 'A'; 161 my @a = map ord, split(/$x/, join("", map chr, (1234, ord($x), 2345))); 162 is("@a", "1234 2345"); 163} 164 165{ 166 # bug id 20000427.003 167 168 use warnings; 169 use strict; 170 171 my $sushi = "\x{b36c}\x{5a8c}\x{ff5b}\x{5079}\x{505b}"; 172 173 my @charlist = split //, $sushi; 174 my $r = ''; 175 foreach my $ch (@charlist) { 176 $r = $r . " " . sprintf "U+%04X", ord($ch); 177 } 178 179 is($r, " U+B36C U+5A8C U+FF5B U+5079 U+505B"); 180} 181 182{ 183 my $s = "\x20\x40\x{80}\x{100}\x{80}\x40\x20"; 184 185 SKIP: { 186 if (ord('A') == 193) { 187 skip("EBCDIC", 1); 188 } else { 189 # bug id 20000426.003 190 191 my ($a, $b, $c) = split(/\x40/, $s); 192 ok($a eq "\x20" && $b eq "\x{80}\x{100}\x{80}" && $c eq $a); 193 } 194 } 195 196 my ($a, $b) = split(/\x{100}/, $s); 197 ok($a eq "\x20\x40\x{80}" && $b eq "\x{80}\x40\x20"); 198 199 my ($a, $b) = split(/\x{80}\x{100}\x{80}/, $s); 200 ok($a eq "\x20\x40" && $b eq "\x40\x20"); 201 202 SKIP: { 203 if (ord('A') == 193) { 204 skip("EBCDIC", 1); 205 } else { 206 my ($a, $b) = split(/\x40\x{80}/, $s); 207 ok($a eq "\x20" && $b eq "\x{100}\x{80}\x40\x20"); 208 } 209 } 210 211 my ($a, $b, $c) = split(/[\x40\x{80}]+/, $s); 212 ok($a eq "\x20" && $b eq "\x{100}" && $c eq "\x20"); 213} 214 215{ 216 # 20001205.014 217 218 my $a = "ABC\x{263A}"; 219 220 my @b = split( //, $a ); 221 222 is(scalar @b, 4); 223 224 ok(length($b[3]) == 1 && $b[3] eq "\x{263A}"); 225 226 $a =~ s/^A/Z/; 227 ok(length($a) == 4 && $a eq "ZBC\x{263A}"); 228} 229 230{ 231 my @a = split(/\xFE/, "\xFF\xFE\xFD"); 232 233 ok(@a == 2 && $a[0] eq "\xFF" && $a[1] eq "\xFD"); 234} 235 236{ 237 # check that PMf_WHITE is cleared after \s+ is used 238 # reported in <20010627113312.RWGY6087.viemta06@localhost> 239 my $r; 240 foreach my $pat ( qr/\s+/, qr/ll/ ) { 241 $r = join ':' => split($pat, "hello cruel world"); 242 } 243 is($r, "he:o cruel world"); 244} 245 246 247{ 248 # split /(A)|B/, "1B2" should return (1, undef, 2) 249 my @x = split /(A)|B/, "1B2"; 250 ok($x[0] eq '1' and (not defined $x[1]) and $x[2] eq '2'); 251} 252 253{ 254 # [perl #17064] 255 my $warn; 256 local $SIG{__WARN__} = sub { $warn = join '', @_; chomp $warn }; 257 my $char = "\x{10f1ff}"; 258 my @a = split /\r?\n/, "$char\n"; 259 ok(@a == 1 && $a[0] eq $char && !defined($warn)); 260} 261 262{ 263 # [perl #18195] 264 for my $u (0, 1) { 265 for my $a (0, 1) { 266 $_ = 'readin,database,readout'; 267 utf8::upgrade $_ if $u; 268 /(.+)/; 269 my @d = split /[,]/,$1; 270 is(join (':',@d), 'readin:database:readout', "[perl #18195]"); 271 } 272 } 273} 274 275{ 276 $p="a,b"; 277 utf8::upgrade $p; 278 eval { @a=split(/[, ]+/,$p) }; 279 is ("$@-@a-", '-a b-', '#20912 - split() to array with /[]+/ and utf8'); 280} 281 282{ 283 is (\@a, \@{"a"}, '@a must be global for following test'); 284 $p=""; 285 $n = @a = split /,/,$p; 286 is ($n, 0, '#21765 - pmreplroot hack used to return undef for 0 iters'); 287} 288 289{ 290 # [perl #28938] 291 # assigning off the end of the array after a split could leave garbage 292 # in the inner elements 293 294 my $x; 295 @a = split /,/, ',,,,,'; 296 $a[3]=1; 297 $x = \$a[2]; 298 is (ref $x, 'SCALAR', '#28938 - garbage after extend'); 299} 300 301