Re: [perl #40641] crash with unicode characters in regex comment

Message-Id: <20061104215302.3325.BQW10602@nifty.com>

p4raw-id: //depot/perl@29204

3 files changed, 281 insertions, 1 deletions

diff --git a/t/uni/greek.t b/t/uni/greek.t
new file mode 100644
index 0000000000..a8102f3880
--- /dev/null
+++ b/t/uni/greek.t
@@ -0,0 +1,119 @@
+BEGIN {
+    if ($ENV{'PERL_CORE'}){
+        chdir 't';
+        @INC = '../lib';
+    }
+    require Config; import Config;
+    if ($Config{'extensions'} !~ /\bEncode\b/) {
+      print "1..0 # Skip: Encode was not built\n";
+      exit 0;
+    }
+    if (ord("A") == 193) {
+        print "1..0 # Skip: EBCDIC\n";
+        exit 0;
+    }
+    unless (PerlIO::Layer->find('perlio')){
+        print "1..0 # Skip: PerlIO required\n";
+        exit 0;
+    }
+    if ($ENV{PERL_CORE_MINITEST}) {
+        print "1..0 # Skip: no dynamic loading on miniperl, no Encode\n";
+        exit 0;
+    }
+    $| = 1;
+    require './test.pl';
+}
+
+plan tests => 72;
+
+use encoding "greek"; # iso 8859-7
+
+# U+0391, \xC1, \301, GREEK CAPITAL LETTER ALPHA
+# U+03B1, \xE1, \341, GREEK SMALL LETTER ALPHA
+
+ok("\xC1"    =~ /\xC1/,     '\xC1 to /\xC1/');
+ok("\x{391}" =~ /\xC1/,     '\x{391} to /\xC1/');
+ok("\xC1"    =~ /\x{C1}/,   '\xC1 to /\x{C1}/');
+ok("\x{391}" =~ /\x{C1}/,   '\x{391} to /\x{C1}/');
+ok("\xC1"    =~ /\301/,     '\xC1 to /\301/');
+ok("\x{391}" =~ /\301/,     '\x{391} to /\301/');
+ok("\xC1"    =~ /\x{391}/,  '\xC1 to /\x{391}/');
+ok("\x{391}" =~ /\x{391}/,  '\x{391} to /\x{391}/');
+
+ok("\xC1"    =~ /\xC1/i,    '\xC1 to /\xC1/i');
+ok("\xE1"    =~ /\xC1/i,    '\xE1 to /\xC1/i');
+ok("\xC1"    =~ /\xE1/i,    '\xC1 to /\xE1/i');
+ok("\xE1"    =~ /\xE1/i,    '\xE1 to /\xE1/i');
+ok("\xC1"    =~ /\x{391}/i, '\xC1 to /\x{391}/i');
+ok("\xE1"    =~ /\x{391}/i, '\xE1 to /\x{391}/i');
+ok("\xC1"    =~ /\x{3B1}/i, '\xC1 to /\x{3B1}/i');
+ok("\xE1"    =~ /\x{3B1}/i, '\xE1 to /\x{3B1}/i');
+
+ok("\xC1"    =~ /[\xC1]/,     '\xC1 to /[\xC1]/');
+ok("\x{391}" =~ /[\xC1]/,     '\x{391} to /[\xC1]/');
+ok("\xC1"    =~ /[\x{C1}]/,   '\xC1 to /[\x{C1}]/');
+ok("\x{391}" =~ /[\x{C1}]/,   '\x{391} to /[\x{C1}]/');
+ok("\xC1"    =~ /[\301]/,     '\xC1 to /[\301]/');
+ok("\x{391}" =~ /[\301]/,     '\x{391} to /[\301]/');
+ok("\xC1"    =~ /[\x{391}]/,  '\xC1 to /[\x{391}]/');
+ok("\x{391}" =~ /[\x{391}]/,  '\x{391} to /[\x{391}]/');
+
+ok("\xC1"    =~ /[\xC1]/i,    '\xC1 to /[\xC1]/i');
+ok("\xE1"    =~ /[\xC1]/i,    '\xE1 to /[\xC1]/i');
+ok("\xC1"    =~ /[\xE1]/i,    '\xC1 to /[\xE1]/i');
+ok("\xE1"    =~ /[\xE1]/i,    '\xE1 to /[\xE1]/i');
+ok("\xC1"    =~ /[\x{391}]/i, '\xC1 to /[\x{391}]/i');
+ok("\xE1"    =~ /[\x{391}]/i, '\xE1 to /[\x{391}]/i');
+ok("\xC1"    =~ /[\x{3B1}]/i, '\xC1 to /[\x{3B1}]/i');
+ok("\xE1"    =~ /[\x{3B1}]/i, '\xE1 to /[\x{3B1}]/i');
+
+ok("\xC1"    =~ '\xC1',       '\xC1 to \'\xC1\'');
+ok("\xC1"    =~ '\x{C1}',     '\xC1 to \'\x{C1}\'');
+ok("\xC1"    =~ '\301',       '\xC1 to \'\301\'');
+ok("\xC1"    =~ '\x{391}',    '\xC1 to \'\x{391}\'');
+ok("\xC1"    =~ '[\xC1]',     '\xC1 to \'[\xC1]\'');
+ok("\xC1"    =~ '[\x{C1}]',   '\xC1 to \'[\x{C1}]\'');
+ok("\xC1"    =~ '[\301]',     '\xC1 to \'[\301]\'');
+ok("\xC1"    =~ '[\x{391}]',  '\xC1 to \'[\x{391}]\'');
+
+ok("\xC1"    =~ /ม/,     '\xC1 to /<ALPHA>/');
+ok("\xE1"    !~ /ม/,     '\xE1 to /<ALPHA>/');
+ok("\xC1"    =~ /ม/i,    '\xC1 to /<ALPHA>/i');
+ok("\xE1"    =~ /ม/i,    '\xE1 to /<ALPHA>/i');
+ok("\xC1"    =~ /[ม]/,   '\xC1 to /[<ALPHA>]/');
+ok("\xE1"    !~ /[ม]/,   '\xE1 to /[<ALPHA>]/');
+ok("\xC1"    =~ /[ม]/i,  '\xC1 to /[<ALPHA>]/i');
+ok("\xE1"    =~ /[ม]/i,  '\xE1 to /[<ALPHA>]/i');
+
+ok("\xC1\xC1"  =~ /ม\xC1/,    '\xC1\xC1 to /<ALPHA>\xC1/');
+ok("\xC1\xC1"  =~ /\xC1ม/,    '\xC1\xC1 to /\xC1<ALPHA>/');
+ok("\xC1\xC1"  =~ /ม\xC1/i,   '\xC1\xC1 to /<ALPHA>\xC1/i');
+ok("\xC1\xC1"  =~ /\xC1ม/i,   '\xC1\xC1 to /\xC1<ALPHA>/i');
+ok("\xC1\xE1"  =~ /ม\xC1/i,   '\xC1\xE1 to /<ALPHA>\xC1/i');
+ok("\xC1\xE1"  =~ /\xC1ม/i,   '\xC1\xE1 to /\xC1<ALPHA>/i');
+ok("\xE1\xE1"  =~ /ม\xC1/i,   '\xE1\xE1 to /<ALPHA>\xC1/i');
+ok("\xE1\xE1"  =~ /\xC1ม/i,   '\xE1\xE1 to /\xC1<ALPHA>/i');
+
+# U+038A, \xBA, GREEK CAPITAL LETTER IOTA WITH TONOS
+# U+03AF, \xDF, GREEK SMALL LETTER IOTA WITH TONOS
+
+ok("\x{38A}"  =~ /\xBA/,      '\x{38A} to /\xBA/');
+ok("\x{38A}"  !~ /\xDF/,      '\x{38A} to /\xDF/');
+ok("\x{38A}"  =~ /\xBA/i,     '\x{38A} to /\xBA/i');
+ok("\x{38A}"  =~ /\xDF/i,     '\x{38A} to /\xDF/i');
+ok("\x{38A}"  =~ /[\xBA]/,    '\x{38A} to /[\xBA]/');
+ok("\x{38A}"  !~ /[\xDF]/,    '\x{38A} to /[\xDF]/');
+ok("\x{38A}"  =~ /[\xBA]/i,   '\x{38A} to /[\xBA]/i');
+ok("\x{38A}"  =~ /[\xDF]/i,   '\x{38A} to /[\xDF]/i');
+
+# \xDF is not LATIN SMALL LETTER SHARP S
+
+ok("SS"   !~ /\xDF/i,   'SS to /\xDF/i');
+ok("Ss"   !~ /\xDF/i,   'Ss to /\xDF/i');
+ok("sS"   !~ /\xDF/i,   'sS to /\xDF/i');
+ok("ss"   !~ /\xDF/i,   'ss to /\xDF/i');
+ok("SS"   !~ /฿/i,      'SS to /<iota-tonos>/i');
+ok("Ss"   !~ /฿/i,      'Ss to /<iota-tonos>/i');
+ok("sS"   !~ /฿/i,      'sS to /<iota-tonos>/i');
+ok("ss"   !~ /฿/i,      'ss to /<iota-tonos>/i');
+
diff --git a/t/uni/latin2.t b/t/uni/latin2.t
new file mode 100644
index 0000000000..08928b6039
--- /dev/null
+++ b/t/uni/latin2.t
@@ -0,0 +1,153 @@
+BEGIN {
+    if ($ENV{'PERL_CORE'}){
+        chdir 't';
+        @INC = '../lib';
+    }
+    require Config; import Config;
+    if ($Config{'extensions'} !~ /\bEncode\b/) {
+      print "1..0 # Skip: Encode was not built\n";
+      exit 0;
+    }
+    if (ord("A") == 193) {
+        print "1..0 # Skip: EBCDIC\n";
+        exit 0;
+    }
+    unless (PerlIO::Layer->find('perlio')){
+        print "1..0 # Skip: PerlIO required\n";
+        exit 0;
+    }
+    if ($ENV{PERL_CORE_MINITEST}) {
+        print "1..0 # Skip: no dynamic loading on miniperl, no Encode\n";
+        exit 0;
+    }
+    $| = 1;
+    require './test.pl';
+}
+
+plan tests => 94;
+
+use encoding "latin2"; # iso 8859-2
+
+# U+00C1, \xC1, \301, LATIN CAPITAL LETTER A WITH ACUTE
+# U+0102, \xC3, \402, LATIN CAPITAL LETTER A WITH BREVE
+# U+00E1, \xE1, \303, LATIN SMALL LETTER A WITH ACUTE
+# U+0103, \xE3, \403, LATIN SMALL LETTER A WITH BREVE
+
+ok("\xC1"    =~ /\xC1/,     '\xC1 to /\xC1/');
+ok("\x{C1}"  =~ /\x{C1}/,   '\x{C1} to /\x{C1}/');
+ok("\xC3"    =~ /\xC3/,     '\xC3 to /\xC3/');
+ok("\x{102}" =~ /\xC3/,     '\x{102} to /\xC3/');
+ok("\xC3"    =~ /\x{C3}/,   '\xC3 to /\x{C3}/');
+ok("\x{102}" =~ /\x{C3}/,   '\x{102} to /\x{C3}/');
+ok("\xC3"    =~ /\x{102}/,  '\xC3 to /\x{102}/');
+ok("\x{102}" =~ /\x{102}/,  '\x{102} to /\x{102}/');
+
+ok("\xC1"    =~ /\xC1/i,    '\xC1 to /\xC1/i');
+ok("\xE1"    =~ /\xC1/i,    '\xE1 to /\xC1/i');
+ok("\xC1"    =~ /\xE1/i,    '\xC1 to /\xE1/i');
+ok("\xE1"    =~ /\xE1/i,    '\xE1 to /\xE1/i');
+ok("\x{102}" =~ /\xC3/i,    '\x{102} to /\xC3/i');
+ok("\x{103}" =~ /\xC3/i,    '\x{103} to /\xC3/i');
+ok("\x{102}" =~ /\xE3/i,    '\x{102} to /\xE3/i');
+ok("\x{103}" =~ /\xE3/i,    '\x{103} to /\xE3/i');
+
+ok("\xC1"    =~ /[\xC1]/,     '\xC1 to /[\xC1]/');
+ok("\x{C1}"  =~ /[\x{C1}]/,   '\x{C1} to /[\x{C1}]/');
+ok("\xC3"    =~ /[\xC3]/,     '\xC3 to /[\xC3]/');
+ok("\x{102}" =~ /[\xC3]/,     '\x{102} to /[\xC3]/');
+ok("\xC3"    =~ /[\x{C3}]/,   '\xC3 to /[\x{C3}]/');
+ok("\x{102}" =~ /[\x{C3}]/,   '\x{102} to /[\x{C3}]/');
+ok("\xC3"    =~ /[\x{102}]/,  '\xC3 to /[\x{102}]/');
+ok("\x{102}" =~ /[\x{102}]/,  '\x{102} to /[\x{102}]/');
+
+ok("\xC1"    =~ /[\xC1]/i,  '\xC1 to /[\xC1]/i');
+ok("\xE1"    =~ /[\xC1]/i,  '\xE1 to /[\xC1]/i');
+ok("\xC1"    =~ /[\xE1]/i,  '\xC1 to /[\xE1]/i');
+ok("\xE1"    =~ /[\xE1]/i,  '\xE1 to /[\xE1]/i');
+ok("\x{102}" =~ /[\xC3]/i,  '\x{102} to /[\xC3]/i');
+ok("\x{103}" =~ /[\xC3]/i,  '\x{103} to /[\xC3]/i');
+ok("\x{102}" =~ /[\xE3]/i,  '\x{102} to /[\xE3]/i');
+ok("\x{103}" =~ /[\xE3]/i,  '\x{103} to /[\xE3]/i');
+
+ok("\xC1"    =~ '\xC1',       '\xC1 to \'\xC1\'');
+ok("\xC1"    =~ '\x{C1}',     '\xC1 to \'\x{C1}\'');
+ok("\xC3"    =~ '\303',       '\xC3 to \'\303\'');
+ok("\xC3"    =~ '\x{102}',    '\xC3 to \'\x{102}\'');
+ok("\xC1"    =~ '[\xC1]',     '\xC1 to \'[\xC1]\'');
+ok("\xC1"    =~ '[\x{C1}]',   '\xC1 to \'[\x{C1}]\'');
+ok("\xC3"    =~ '[\303]',     '\xC3 to \'[\303]\'');
+ok("\xC3"    =~ '[\x{102}]',  '\xC3 to \'[\x{102}]\'');
+
+ok("\xC1"    =~ /ม/,     '\xC1 to /<A-acute>/');
+ok("\xE1"    !~ /ม/,     '\xE1 to /<A-acute>/');
+ok("\xC1"    =~ /ม/i,    '\xC1 to /<A-acute>/i');
+ok("\xE1"    =~ /ม/i,    '\xE1 to /<A-acute>/i');
+ok("\xC1"    =~ /[ม]/,   '\xC1 to /[<A-acute>]/');
+ok("\xE1"    !~ /[ม]/,   '\xE1 to /[<A-acute>]/');
+ok("\xC1"    =~ /[ม]/i,  '\xC1 to /[<A-acute>]/i');
+ok("\xE1"    =~ /[ม]/i,  '\xE1 to /[<A-acute>]/i');
+
+ok("\xC1\xC1"  =~ /ม\xC1/,    '\xC1\xC1 to /<A-acute>\xC1/');
+ok("\xC1\xC1"  =~ /\xC1ม/,    '\xC1\xC1 to /\xC1<A-acute>/');
+ok("\xC1\xC1"  =~ /ม\xC1/i,   '\xC1\xC1 to /<A-acute>\xC1/i');
+ok("\xC1\xC1"  =~ /\xC1ม/i,   '\xC1\xC1 to /\xC1<A-acute>/i');
+ok("\xC1\xE1"  =~ /ม\xC1/i,   '\xC1\xE1 to /<A-acute>\xC1/i');
+ok("\xC1\xE1"  =~ /\xC1ม/i,   '\xC1\xE1 to /\xC1<A-acute>/i');
+ok("\xE1\xE1"  =~ /ม\xC1/i,   '\xE1\xE1 to /<A-acute>\xC1/i');
+ok("\xE1\xE1"  =~ /\xC1ม/i,   '\xE1\xE1 to /\xC1<A-acute>/i');
+
+# \xDF is LATIN SMALL LETTER SHARP S
+
+ok("\xDF" =~ /\xDF/,    '\xDF to /\xDF/');
+ok("\xDF" =~ /\xDF/i,   '\xDF to /\xDF/i');
+ok("\xDF" =~ /[\xDF]/,  '\xDF to /[\xDF]/');
+ok("\xDF" =~ /[\xDF]/i, '\xDF to /[\xDF]/i');
+ok("\xDF" =~ /฿/,       '\xDF to /<sharp-s>/');
+ok("\xDF" =~ /฿/i,      '\xDF to /<sharp-s>/i');
+ok("\xDF" =~ /[฿]/,     '\xDF to /[<sharp-s>]/');
+ok("\xDF" =~ /[฿]/i,    '\xDF to /[<sharp-s>]/i');
+
+ok("SS"   =~ /\xDF/i,   'SS to /\xDF/i');
+ok("Ss"   =~ /\xDF/i,   'Ss to /\xDF/i');
+ok("sS"   =~ /\xDF/i,   'sS to /\xDF/i');
+ok("ss"   =~ /\xDF/i,   'ss to /\xDF/i');
+ok("SS"   =~ /฿/i,      'SS to /<sharp-s>/i');
+ok("Ss"   =~ /฿/i,      'Ss to /<sharp-s>/i');
+ok("sS"   =~ /฿/i,      'sS to /<sharp-s>/i');
+ok("ss"   =~ /฿/i,      'ss to /<sharp-s>/i');
+
+ok("\xC3" =~ /\303/,     '\xC1 to /\303/');
+ok("\303" =~ /\303/,     '\303 to /\303/');
+ok("\xC3" =~ /\303/i,    '\xC1 to /\303/i');
+ok("\xE3" =~ /\303/i,    '\xC1 to /\303/i');
+ok("\xC3" =~ /[\303]/,   '\xC1 to /[\303]/');
+ok("\303" =~ /[\303]/,   '\303 to /[\303]/');
+ok("\xC3" =~ /[\303]/i,  '\xC1 to /[\303]/i');
+ok("\xE3" =~ /[\303]/i,  '\xC1 to /[\303]/i');
+
+ok("\xC3" =~ /\402/,     '\xC1 to /\402/');
+ok("\402" =~ /\402/,     '\402 to /\402/');
+ok("\xC3" =~ /\402/i,    '\xC1 to /\402/i');
+ok("\xE3" =~ /\402/i,    '\xC1 to /\402/i');
+ok("\xC3" =~ /[\402]/,   '\xC1 to /[\402]/');
+ok("\402" =~ /[\402]/,   '\402 to /[\402]/');
+ok("\xC3" =~ /[\402]/i,  '\xC1 to /[\402]/i');
+ok("\xE3" =~ /[\402]/i,  '\xC1 to /[\402]/i');
+
+{
+    my $re = '(?i:\xC1)';
+
+    ok("\xC1" =~ $re, '\xC1 to (?i:\xC1)');
+    ok("\xE1" =~ $re, '\xE1 to (?i:\xC1)');
+
+    utf8::downgrade($re);
+
+    ok("\xC1" =~ $re, '\xC1 to (?i:\xC1) down');
+    ok("\xE1" =~ $re, '\xE1 to (?i:\xC1) down');
+
+    utf8::upgrade($re);
+
+    ok("\xC1" =~ $re, '\xC1 to (?i:\xC1) up');
+    ok("\xE1" =~ $re, '\xE1 to (?i:\xC1) up');
+}
+
diff --git a/t/uni/tr_utf8.t b/t/uni/tr_utf8.t
index 606a84a9c0..354156a641 100644
--- a/t/uni/tr_utf8.t
+++ b/t/uni/tr_utf8.t
@@ -31,7 +31,7 @@ BEGIN {
 }
 
 use strict;
-use Test::More tests => 7;
+use Test::More tests => 8;
 
 use encoding 'utf8';
 
@@ -67,4 +67,12 @@ is($str, $hiragana, "s/// # hiragana -> katakana");
   $line =~ tr/bcdeghijklmnprstvwxyz$02578/ื‘ืฆื“ืขื’ื”ื™ืฒืงืœืžื ืคึผืจืกื˜ืฐืฉื›ื™ื–ืฉืฑืชืฒื—ื/;
   is($line, "aื‘ืฆื“ืขfื’ื”ื™ืฒืงืœืžื oืคqึผืจืกuื˜ืฐืฉื›ื™ื–ืฉ1ืฑ34ืช6ืฒื—9", "[perl #16843]");
 }
+
+{
+  # [perl #40641]
+  my $str = qq/Gebรครครครครครครครครครครครครครครครครครครครคude/;
+  my $reg = qr/Gebรครครครครครครครครครครครครครครครครครครครคude/;
+  ok($str =~ /$reg/, "[perl #40641]");
+}
+
 __END__