test.pl

use strict;
use warnings;

use utf8;

use Test::More;
use Test::Builder;

BEGIN {
    use_ok('AvoidCoreTextBug');
}

my $Tester = Test::Builder->new;

subtest simple => sub {
    is_filterd("\x{600}\x{20}");
    is_filterd("&#x600\x{20}");
    is_filterd("؀\x{20}");

    is_filterd("\x{600} ");
    is_filterd("&#x600 ");
    is_filterd("؀ ");

    is_filterd("\x{600} ");
    is_filterd("&#x600 ");
    is_filterd("؀ ");

    is_filterd("\x{600} ");
    is_filterd("&#x600&#x20");
    is_filterd("؀&#32");
};

subtest 'simple no entity reference' => sub {
    is_filterd("\x{600}\x{20}", 1);
    is_no_filterd("&#x600\x{20}", 1);
    is_no_filterd("؀\x{20}", 1);

    is_filterd("\x{600} ", 1);
    is_no_filterd("&#x600 ", 1);
    is_no_filterd("؀ ", 1);

    is_no_filterd("\x{600} ", 1);
    is_no_filterd("&#x600 ", 1);
    is_no_filterd("؀ ", 1);

    is_no_filterd("\x{600} ", 1);
    is_no_filterd("&#x600&#x20", 1);
    is_no_filterd("؀&#32", 1);
};

subtest 'multi line' => sub {
    my $filterd = AvoidCoreTextBug::filter("\na\nb\x{600}\x{20}a\nb\nc\n\x{600}\x{20}\n\n\n");
    is($filterd, "\na\nb\x{600}\x{20}\x{200e}a\nb\nc\n\x{600}\x{20}\x{200e}\n\n\n");
};

subtest ascii => sub {
    is_no_filterd("abc");
};

subtest with_tag => sub {
    is_no_filterd("foo<a>bar");
    is_no_filterd("foo <a> bar </a> baz");
    is_no_filterd("foo <a> &nbsp;");
};

subtest with_entity => sub {
    is_no_filterd("a&#98;");
    is_no_filterd("a&#98;&#x63");
    is_no_filterd("a&#98;&#x63;");
    is_no_filterd("a&#98;&#x63;d");
};

subtest 'with 0x00-0xff entity' => sub {
    is_no_filterd("a&lt;&gt;b");
    is_no_filterd("a&#x3c;&#x3e;b");
    is_no_filterd("a&#60;&#62;b");
    is_no_filterd("&#X0000003C;&#x00003E");
    is_no_filterd("&#000000060;&#0000062");
    is_no_filterd("&#000000060");
};

subtest 'with 0x100- entity (not Arabic)' => sub {
    is_no_filterd("&#x3042;&#x3044&#x0003046;&#12360;&#00012362;");
    is_no_filterd("a&#x3042;i&#x3044u&#x0003046;e&#12360;o&#00012362;");
    is_no_filterd("&#97;&#x3042;&#x62;&#x3044u&#x0003046;e&#12360;o&#00012362;");
    is_no_filterd("a&#x3042;b&#x62;");
};

subtest with_Arabic => sub {
    is_no_filterd("&#x600;");
    is_filterd("&#x600;\t");
    is_no_filterd("&#x750;");
    is_filterd("&#x750&nbsp");
    is_no_filterd("&#x8a0");
    is_filterd("&#x8a0 ");
    is_no_filterd("&#xfb50;");
    is_filterd("&#xfb50;\n");
    is_no_filterd("&#x7e70;");
    is_filterd("&#x7e70;\r");
    is_no_filterd("&#x10e60;");
    is_filterd("&#x10e60;\x{07}");
    is_no_filterd("&#x1ee00;");
    is_filterd("&#x1ee00; ");

    is_no_filterd("&#no_x200f;");
    is_filterd("&#x200f; ");
    is_no_filterd("&#x202b;");
    is_filterd("&#x202b; ");
    is_no_filterd("&#x202e;");
    is_filterd("&#x202e; ");
};

subtest with_Hebrew => sub {
    is_no_filterd("&#x590;");
    is_filterd("&#x590;&nbsp;");
    is_no_filterd("&#xFB1D;");
    is_filterd("&#xFB1D;\t");
};

subtest with_Syriac => sub {
    is_filterd("&#x700; ");
};

subtest with_all_right_to_left_char => sub {
    foreach my $code (0x600 .. 0x6ff,
                      0x750 .. 0x77f,
                      0x8a0 .. 0x8ff,
                      0xfb50 .. 0xfdff,
                      0x7e70 .. 0xfeff,
                      0x10e60 .. 0x10e7f,
                      0x1ee00 .. 0x1eeff,
                      0x590 .. 0x05FF,
                      0xfb1d .. 0xfb4f,
                      0x700 .. 0x74f){
        my $str = "&#$code";
        my $filterd = AvoidCoreTextBug::filter($str);

        if($str ne $filterd){
            is($str, $filterd, $code);
        }
        $filterd = AvoidCoreTextBug::filter($str . " ");
        my $expected = $str . " \x{200e}";
        if($expected ne $filterd){
            is($str, $filterd, "$code + space");
        }
    }
    ok(1);
};

subtest separator => sub {
    is_filterd("&#x600;\x{2003}");
    foreach my $code (0x20, 0xa0,
                      0x1680, 0x180e,
                      0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, 0x2006, 0x2007, 0x2008, 0x2009, 0x200a,
                      0x202f, 0x205f,
                      0x3000){
    }
};

subtest complex => sub {
    is_no_filterd("a&#x600;b<&lt;&#1234あいう&rlm;&#xfb50か&#x200fきく");
    is(AvoidCoreTextBug::filter("a&#x600\tb<&lt;&#1234あいう&rlm; &#xfb50&nbspか&#x200f\nきく"), "a&#x600\t\x{200e}b<&lt;&#1234あいう&rlm; \x{200e}&#xfb50&nbsp\x{200e}か&#x200f\n\x{200e}きく");
};

subtest 'bug fix' => sub {
    is_no_filterd("}&nbsp;");
    is_filterd("\x{600}&shy;");
    is_filterd("\x{600}&#173;");
    is_filterd("\x{600}\x{ad}");
};


sub is_filterd {
    my($original, $dont_convert_entity_reference) = @_;

    my $filterd = AvoidCoreTextBug::filter($original, $dont_convert_entity_reference);
    $Tester->is_eq($filterd, $original . "\x{200e}");
}

sub is_no_filterd {
    my($original, $dont_convert_entity_reference) = @_;

    my $filterd = AvoidCoreTextBug::filter($original, $dont_convert_entity_reference);
    $Tester->is_eq($filterd, $original);
}

done_testing;