$| = 1;
-print "1..615\n";
+print "1..683\n";
BEGIN {
chdir 't' if -d 't';
@INC = '../lib';
}
-use re 'asciirange'; # Compute ranges in ASCII space
-
eval 'use Config'; # Defaults assumed if this fails
$x = "abc\ndef\n";
print "ok $test\n";
$test++;
+my $ordA = ord('A');
+
$_ = "a\x{100}b";
if (/(.)(\C)(\C)(.)/) {
print "ok 232\n";
} else {
print "not ok 233\n";
}
- if ($2 eq "\xC4") {
- print "ok 234\n";
- } else {
- print "not ok 234\n";
- }
- if ($3 eq "\x80") {
- print "ok 235\n";
+ if ($ordA == 65) { # ASCII (or equivalent), should be UTF-8
+ if ($2 eq "\xC4") {
+ print "ok 234\n";
+ } else {
+ print "not ok 234\n";
+ }
+ if ($3 eq "\x80") {
+ print "ok 235\n";
+ } else {
+ print "not ok 235\n";
+ }
+ } elsif ($ordA == 193) { # EBCDIC (or equivalent), should be UTF-EBCDIC
+ if ($2 eq "\x8C") {
+ print "ok 234\n";
+ } else {
+ print "not ok 234\n";
+ }
+ if ($3 eq "\x41") {
+ print "ok 235\n";
+ } else {
+ print "not ok 235\n";
+ }
} else {
- print "not ok 235\n";
+ for (234..235) {
+ print "not ok $_ # ord('A') == $ordA\n";
+ }
}
if ($4 eq "b") {
print "ok 236\n";
if (/(\C)/g) {
print "ok 237\n";
# currently \C are still tagged as UTF-8
- if ($1 eq "\xC4") {
- print "ok 238\n";
+ if ($ordA == 65) {
+ if ($1 eq "\xC4") {
+ print "ok 238\n";
+ } else {
+ print "not ok 238\n";
+ }
+ } elsif ($ordA == 193) {
+ if ($1 eq "\x8C") {
+ print "ok 238\n";
+ } else {
+ print "not ok 238\n";
+ }
} else {
- print "not ok 238\n";
+ print "not ok 238 # ord('A') == $ordA\n";
}
} else {
for (237..238) {
if (/(\C)/g) {
print "ok 239\n";
# currently \C are still tagged as UTF-8
- if ($1 eq "\x80") {
- print "ok 240\n";
+ if ($ordA == 65) {
+ if ($1 eq "\x80") {
+ print "ok 240\n";
+ } else {
+ print "not ok 240\n";
+ }
+ } elsif ($ordA == 193) {
+ if ($1 eq "\x41") {
+ print "ok 240\n";
+ } else {
+ print "not ok 240\n";
+ }
} else {
- print "not ok 240\n";
+ print "not ok 240 # ord('A') == $ordA\n";
}
} else {
for (239..240) {
"#latin[$latin]\nnot ok $test\n";
$test++;
$latin =~ s/stra\337e/straße/; # \303\237 after the 2nd a
- use utf8;
+ use utf8; # needed for the raw UTF-8
$latin =~ s!(s)tr(?:aß|s+e)!$1tr.!; # \303\237 after the a
}
}
print "ok 615\n";
}
+{
+ # from japhy
+ my $w;
+ use warnings;
+ local $SIG{__WARN__} = sub { $w .= shift };
+
+ $w = "";
+ eval 'qr/(?c)/';
+ print "not " if $w !~ /^Useless \(\?c\)/;
+ print "ok 616\n";
+
+ $w = "";
+ eval 'qr/(?-c)/';
+ print "not " if $w !~ /^Useless \(\?-c\)/;
+ print "ok 617\n";
+
+ $w = "";
+ eval 'qr/(?g)/';
+ print "not " if $w !~ /^Useless \(\?g\)/;
+ print "ok 618\n";
+
+ $w = "";
+ eval 'qr/(?-g)/';
+ print "not " if $w !~ /^Useless \(\?-g\)/;
+ print "ok 619\n";
+
+ $w = "";
+ eval 'qr/(?o)/';
+ print "not " if $w !~ /^Useless \(\?o\)/;
+ print "ok 620\n";
+
+ $w = "";
+ eval 'qr/(?-o)/';
+ print "not " if $w !~ /^Useless \(\?-o\)/;
+ print "ok 621\n";
+
+ # now test multi-error regexes
+
+ $w = "";
+ eval 'qr/(?g-o)/';
+ print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-o\)/;
+ print "ok 622\n";
+
+ $w = "";
+ eval 'qr/(?g-c)/';
+ print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-c\)/;
+ print "ok 623\n";
+
+ $w = "";
+ eval 'qr/(?o-cg)/'; # (?c) means (?g) error won't be thrown
+ print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?-c\)/;
+ print "ok 624\n";
+
+ $w = "";
+ eval 'qr/(?ogc)/';
+ print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?g\).*\nUseless \(\?c\)/;
+ print "ok 625\n";
+}
+
+# More Unicode "class" tests
+
+{
+ use charnames ':full';
+
+ print "not " unless "\N{LATIN CAPITAL LETTER A}" =~ /\p{InBasicLatin}/;
+ print "ok 626\n";
+
+ print "not " unless "\N{LATIN CAPITAL LETTER A WITH GRAVE}" =~ /\p{InLatin1Supplement}/;
+ print "ok 627\n";
+
+ print "not " unless "\N{LATIN CAPITAL LETTER A WITH MACRON}" =~ /\p{InLatinExtendedA}/;
+ print "ok 628\n";
+
+ print "not " unless "\N{LATIN SMALL LETTER B WITH STROKE}" =~ /\p{InLatinExtendedB}/;
+ print "ok 629\n";
+
+ print "not " unless "\N{KATAKANA LETTER SMALL A}" =~ /\p{InKatakana}/;
+ print "ok 630\n";
+}
+
+$_ = "foo";
+
+eval <<"EOT"; die if $@;
+ /f
+ o\r
+ o
+ \$
+ /x && print "ok 631\n";
+EOT
+
+eval <<"EOT"; die if $@;
+ /f
+ o
+ o
+ \$\r
+ /x && print "ok 632\n";
+EOT
+
+#test /o feature
+sub test_o { $_[0] =~/$_[1]/o; return $1}
+if(test_o('abc','(.)..') eq 'a') {
+ print "ok 633\n";
+} else {
+ print "not ok 633\n";
+}
+if(test_o('abc','..(.)') eq 'a') {
+ print "ok 634\n";
+} else {
+ print "not ok 634\n";
+}
+
+# 635..639: ID 20010619.003 (only the space character is
+# supposed to be [:print:], not the whole isprint()).
+
+print "not " if "\n" =~ /[[:print:]]/;
+print "ok 635\n";
+
+print "not " if "\t" =~ /[[:print:]]/;
+print "ok 636\n";
+
+# Amazingly vertical tabulator is the same in ASCII and EBCDIC.
+print "not " if "\014" =~ /[[:print:]]/;
+print "ok 637\n";
+
+print "not " if "\r" =~ /[[:print:]]/;
+print "ok 638\n";
+
+print "not " unless " " =~ /[[:print:]]/;
+print "ok 639\n";
+
+##
+## Test basic $^N usage outside of a regex
+##
+$x = "abcdef";
+$T="ok 640\n";if ($x =~ /cde/ and not defined $^N) {print $T} else {print "not $T"};
+$T="ok 641\n";if ($x =~ /(cde)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 642\n";if ($x =~ /(c)(d)(e)/ and $^N eq "e") {print $T} else {print "not $T"};
+$T="ok 643\n";if ($x =~ /(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 644\n";if ($x =~ /(foo)|(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 645\n";if ($x =~ /(c(d)e)|(foo)/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 646\n";if ($x =~ /(c(d)e)|(abc)/ and $^N eq "abc") {print $T} else {print "not $T"};
+$T="ok 647\n";if ($x =~ /(c(d)e)|(abc)x/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 648\n";if ($x =~ /(c(d)e)(abc)?/ and $^N eq "cde") {print $T} else {print "not $T"};
+$T="ok 649\n";if ($x =~ /(?:c(d)e)/ and $^N eq "d" ) {print $T} else {print "not $T"};
+$T="ok 650\n";if ($x =~ /(?:c(d)e)(?:f)/ and $^N eq "d" ) {print $T} else {print "not $T"};
+$T="ok 651\n";if ($x =~ /(?:([abc])|([def]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
+$T="ok 652\n";if ($x =~ /(?:([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
+$T="ok 653\n";if ($x =~ /(([ace])|([bd]))*/ and $^N eq "e" ){print $T} else {print "not $T"};
+{
+ $T="ok 654\n";if($x =~ /(([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
+}
+## test to see if $^N is automatically localized -- it should now
+## have the value set in test 653
+$T="ok 655\n";if ($^N eq "e" ){print $T} else {print "not $T"};
+
+##
+## Now test inside (?{...})
+##
+$T="ok 656\n";if ($x =~ /a([abc])(?{$y=$^N})c/ and $y eq "b" ){print $T} else {print "not $T"};
+$T="ok 657\n";if ($x =~ /a([abc]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
+$T="ok 658\n";if ($x =~ /a([abcdefg]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
+$T="ok 659\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})d)(?{$z=$^N})e/ and $y eq "bc" and $z eq "abcd")
+ {print $T} else {print "not $T"};
+$T="ok 660\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})de)(?{$z=$^N})/ and $y eq "bc" and $z eq "abcde")
+ {print $T} else {print "not $T"};
+
+# Test the Unicode script classes
+
+print "not " unless chr(0x100) =~ /\p{InLatin}/; # outside Latin-1
+print "ok 661\n";
+
+print "not " unless chr(0x212b) =~ /\p{InLatin}/; # Angstrom sign, very outside
+print "ok 662\n";
+
+print "not " unless chr(0x5d0) =~ /\p{InHebrew}/; # inside HebrewBlock
+print "ok 663\n";
+
+print "not " unless chr(0xfb4f) =~ /\p{InHebrew}/; # outside HebrewBlock
+print "ok 664\n";
+
+print "not " unless chr(0xb5) =~ /\p{InGreek}/; # singleton (not in a range)
+print "ok 665\n";
+
+print "not " unless chr(0x37a) =~ /\p{InGreek}/; # singleton
+print "ok 666\n";
+
+print "not " unless chr(0x386) =~ /\p{InGreek}/; # singleton
+print "ok 667\n";
+
+print "not " unless chr(0x387) =~ /\P{InGreek}/; # not there
+print "ok 668\n";
+
+print "not " unless chr(0x388) =~ /\p{InGreek}/; # range
+print "ok 669\n";
+
+print "not " unless chr(0x38a) =~ /\p{InGreek}/; # range
+print "ok 670\n";
+
+print "not " unless chr(0x38b) =~ /\P{InGreek}/; # not there
+print "ok 671\n";
+
+print "not " unless chr(0x38c) =~ /\p{InGreek}/; # singleton
+print "ok 672\n";
+
+##
+## Test [:cntrl:]...
+##
+## Should probably put in tests for all the POSIX stuff, but not sure how to
+## guarantee a specific locale......
+##
+$AllBytes = join('', map { chr($_) } 0..255);
+($x = $AllBytes) =~ s/[[:cntrl:]]//g;
+if ($x ne join('', map { chr($_) } 0x20..0x7E, 0x80..0xFF)) { print "not " };
+print "ok 673\n";
+
+($x = $AllBytes) =~ s/[^[:cntrl:]]//g;
+if ($x ne join('', map { chr($_) } 0..0x1F, 0x7F)) { print "not " };
+print "ok 674\n";
+
+# With /s modifier UTF8 chars were interpreted as bytes
+{
+ my $a = "Hello \x{263A} World";
+
+ my @a = ($a =~ /./gs);
+
+ print "not " unless $#a == 12;
+ print "ok 675\n";
+}
+
+@a = ("foo\nbar" =~ /./g);
+print "ok 676\n" if @a == 6 && "@a" eq "f o o b a r";
+
+@a = ("foo\nbar" =~ /./gs);
+print "ok 677\n" if @a == 7 && "@a" eq "f o o \n b a r";
+
+@a = ("foo\nbar" =~ /\C/g);
+print "ok 678\n" if @a == 7 && "@a" eq "f o o \n b a r";
+
+@a = ("foo\nbar" =~ /\C/gs);
+print "ok 679\n" if @a == 7 && "@a" eq "f o o \n b a r";
+
+@a = ("foo\n\x{100}bar" =~ /./g);
+print "ok 680\n" if @a == 7 && "@a" eq "f o o \x{100} b a r";
+
+@a = ("foo\n\x{100}bar" =~ /./gs);
+print "ok 681\n" if @a == 8 && "@a" eq "f o o \n \x{100} b a r";
+
+($a, $b) = map { chr } ord('A') == 65 ? (0xc4, 0x80) : (0x8c, 0x41);
+
+@a = ("foo\n\x{100}bar" =~ /\C/g);
+print "ok 682\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
+
+@a = ("foo\n\x{100}bar" =~ /\C/gs);
+print "ok 683\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
+