15 ($a,$b,$c) = split($FS,$_);
17 is(join(';',$a,$b,$c), 'a;b;c');
20 is(join("$_",@ary), 'aa:b:cc');
23 my @xyz = (@ary = split(//));
24 is(join(".",@ary), "a.b.c.\n");
28 is(join(".",@ary), "a.b.c");
30 $_ = join(':',split(' '," a b\tc \t d "));
33 $_ = join(':',split(/ */,"foo bar bie\tdoll"));
34 is($_ , "f:o:o:b:a:r:b:i:e:\t:d:o:l:l");
36 $_ = join(':', 'foo', split(/ /,'a b c'), 'bar');
37 is($_, "foo:a:b::c:bar");
39 # Can we say how many fields to split to?
40 $_ = join(':', split(' ','1 2 3 4 5 6', 3));
41 is($_, '1:2:3 4 5 6');
43 # Can we do it as a variable?
45 $_ = join(':', split(' ','1 2 3 4 5 6', $x));
46 is($_, '1:2:3:4 5 6');
48 # Does the 999 suppress null field chopping?
49 $_ = join(':', split(/:/,'1:2:3:4:5:6:::', 999));
50 is($_ , '1:2:3:4:5:6:::');
52 # Does assignment to a list imply split to one more field than that?
53 $foo = runperl( switches => ['-Dt'], stderr => 1, prog => '($a,$b)=split;' );
54 ok($foo =~ /DEBUGGING/ || $foo =~ /const\n?\Q(IV(3))\E/);
56 # Can we say how many fields to split to when assigning to a list?
57 ($a,$b) = split(' ','1 2 3 4 5 6', 2);
59 is($_, '1:2 3 4 5 6');
61 # do subpatterns generate additional fields (without trailing nulls)?
62 $_ = join '|', split(/,|(-)/, "1-10,20,,,");
65 # do subpatterns generate additional fields (with a limit)?
66 $_ = join '|', split(/,|(-)/, "1-10,20,,,", 10);
67 is($_, "1|-|10||20||||||");
69 # is the 'two undefs' bug fixed?
70 (undef, $a, undef, $b) = qw(1 2 3 4);
75 local(undef, $a, undef, $b) = qw(1 2 3 4);
79 # check splitting of null string
80 $_ = join('|', split(/x/, '',-1), 'Z');
83 $_ = join('|', split(/x/, '', 1), 'Z');
86 $_ = join('|', split(/(p+)/,'',-1), 'Z');
89 $_ = join('|', split(/.?/, '',-1), 'Z');
93 # Are /^/m patterns scanned?
94 $_ = join '|', split(/^a/m, "a b a\na d a", 20);
95 is($_, "| b a\n| d a");
97 # Are /$/m patterns scanned?
98 $_ = join '|', split(/a$/m, "a b a\na d a", 20);
99 is($_, "a b |\na d |");
101 # Are /^/m patterns scanned?
102 $_ = join '|', split(/^aa/m, "aa b aa\naa d aa", 20);
103 is($_, "| b aa\n| d aa");
105 # Are /$/m patterns scanned?
106 $_ = join '|', split(/aa$/m, "aa b aa\naa d aa", 20);
107 is($_, "aa b |\naa d |");
111 @ary = split(/\s*:\s*/);
112 is(($res = join(".",@ary)), "a.b.c.d", $res);
114 # use of match result as pattern (!)
115 is('p:q:r:s', join ':', split('abc' =~ /b/, 'p1q1r1s'));
117 # /^/ treated as /^/m
118 $_ = join ':', split /^/, "ab\ncd\nef\n";
119 is($_, "ab\n:cd\n:ef\n");
121 # see if @a = @b = split(...) optimization works
122 @list1 = @list2 = split ('p',"a p b c p");
123 ok(@list1 == @list2 &&
124 "@list1" eq "@list2" &&
126 "@list1" eq "a b c ");
128 # zero-width assertion
129 $_ = join ':', split /(?=\w)/, "rm b";
134 @ary = map {ord} split //, v1.20.300.4000.50000.4000.300.20.1;
135 is("@ary", "1 20 300 4000 50000 4000 300 20 1");
137 @ary = split(/\x{FE}/, "\x{FF}\x{FE}\x{FD}"); # bug id 20010105.016
139 $ary[0] eq "\xFF" && $ary[1] eq "\xFD" &&
140 $ary[0] eq "\x{FF}" && $ary[1] eq "\x{FD}");
142 @ary = split(/(\x{FE}\xFE)/, "\xFF\x{FF}\xFE\x{FE}\xFD\x{FD}"); # variant of 31
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 $ary[2] eq "\x{FD}\x{FD}");
155 my @a = map ord, split(//, join("", map chr, (1234, 123, 2345)));
156 is("@a", "1234 123 2345");
161 my @a = map ord, split(/$x/, join("", map chr, (1234, ord($x), 2345)));
162 is("@a", "1234 2345");
166 # bug id 20000427.003
171 my $sushi = "\x{b36c}\x{5a8c}\x{ff5b}\x{5079}\x{505b}";
173 my @charlist = split //, $sushi;
175 foreach my $ch (@charlist) {
176 $r = $r . " " . sprintf "U+%04X", ord($ch);
179 is($r, " U+B36C U+5A8C U+FF5B U+5079 U+505B");
183 my $s = "\x20\x40\x{80}\x{100}\x{80}\x40\x20";
186 if (ord('A') == 193) {
189 # bug id 20000426.003
191 my ($a, $b, $c) = split(/\x40/, $s);
192 ok($a eq "\x20" && $b eq "\x{80}\x{100}\x{80}" && $c eq $a);
196 my ($a, $b) = split(/\x{100}/, $s);
197 ok($a eq "\x20\x40\x{80}" && $b eq "\x{80}\x40\x20");
199 my ($a, $b) = split(/\x{80}\x{100}\x{80}/, $s);
200 ok($a eq "\x20\x40" && $b eq "\x40\x20");
203 if (ord('A') == 193) {
206 my ($a, $b) = split(/\x40\x{80}/, $s);
207 ok($a eq "\x20" && $b eq "\x{100}\x{80}\x40\x20");
211 my ($a, $b, $c) = split(/[\x40\x{80}]+/, $s);
212 ok($a eq "\x20" && $b eq "\x{100}" && $c eq "\x20");
218 my $a = "ABC\x{263A}";
220 my @b = split( //, $a );
224 ok(length($b[3]) == 1 && $b[3] eq "\x{263A}");
227 ok(length($a) == 4 && $a eq "ZBC\x{263A}");
231 my @a = split(/\xFE/, "\xFF\xFE\xFD");
233 ok(@a == 2 && $a[0] eq "\xFF" && $a[1] eq "\xFD");
237 # check that PMf_WHITE is cleared after \s+ is used
238 # reported in <20010627113312.RWGY6087.viemta06@localhost>
240 foreach my $pat ( qr/\s+/, qr/ll/ ) {
241 $r = join ':' => split($pat, "hello cruel world");
243 is($r, "he:o cruel world");
248 # split /(A)|B/, "1B2" should return (1, undef, 2)
249 my @x = split /(A)|B/, "1B2";
250 ok($x[0] eq '1' and (not defined $x[1]) and $x[2] eq '2');
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));
266 $_ = 'readin,database,readout';
267 utf8::upgrade $_ if $u;
269 my @d = split /[,]/,$1;
270 is(join (':',@d), 'readin:database:readout', "[perl #18195]");
278 eval { @a=split(/[, ]+/,$p) };
279 is ("$@-@a-", '-a b-', '#20912 - split() to array with /[]+/ and utf8');
283 is (\@a, \@{"a"}, '@a must be global for following test');
285 $n = @a = split /,/,$p;
286 is ($n, 0, '#21765 - pmreplroot hack used to return undef for 0 iters');
291 # assigning off the end of the array after a split could leave garbage
292 # in the inner elements
295 @a = split /,/, ',,,,,';
298 is (ref $x, 'SCALAR', '#28938 - garbage after extend');
301 # check the special casing of split /\s/ and unicode
302 use charnames qw(:full);
303 # below test data is extracted from
305 # Date: 2006-06-07, 23:22:52 GMT [MD]
307 # Unicode Character Database
308 # Copyright (c) 1991-2006 Unicode, Inc.
309 # For terms of use, see http://www.unicode.org/terms_of_use.html
310 # For documentation, see UCD.html
312 ord("\t"), # Cc <control-0009>
313 ord("\n"), # Cc <control-000A>
314 # not PerlSpace # Cc <control-000B>
315 ord("\f"), # Cc <control-000C>
316 ord("\r"), # Cc <control-000D>
318 ord("\N{NEL}"), # Cc <control-0085>
319 ord("\N{NO-BREAK SPACE}"),
321 0x1680, # Zs OGHAM SPACE MARK
322 0x180E, # Zs MONGOLIAN VOWEL SEPARATOR
323 0x2000..0x200A, # Zs [11] EN QUAD..HAIR SPACE
324 0x2028, # Zl LINE SEPARATOR
325 0x2029, # Zp PARAGRAPH SEPARATOR
326 0x202F, # Zs NARROW NO-BREAK SPACE
327 0x205F, # Zs MEDIUM MATHEMATICAL SPACE
328 0x3000 # Zs IDEOGRAPHIC SPACE
330 #diag "Have @{[0+@spaces]} to test\n";
331 foreach my $cp (@spaces) {
332 my $msg = sprintf "Space: U+%04x", $cp;
333 my $space = chr($cp);
334 my $str="A:$space:B\x{FFFD}";
337 my @res=split(/\s+/,$str);
338 ok(@res == 2 && join('-',@res) eq "A:-:B", "$msg - /\\s+/");
340 my $s2 = "$space$space:A:$space$space:B\x{FFFD}";
343 my @r2 = split(' ',$s2);
344 ok(@r2 == 2 && join('-', @r2) eq ":A:-:B", "$msg - ' '");
346 my @r3 = split(/\s+/, $s2);
347 ok(@r3 == 3 && join('-', @r3) eq "-:A:-:B", "$msg - /\\s+/ No.2");