// SKIP: {
// skip( "match()/matched() return undef in this version of perl ($])", 8 ) if $PERL_VERSION_5_005;
ok(!$re->match('foo'), 'not match pattern-6 foo');
$target = 'cat';
ok($re->match($target), "match pattern-6 {$target}");
is($re->matched, '^cat', "match pattern-6 {$target} re");
$target = 'CATFOOD';
ok($re->match($target), "match pattern-6 {$target}");
is($re->matched, '^cat', "match pattern-6 {$target} re");
$target = 'candle';
ok($re->match($target), "match pattern-6 {$target}");
is($re->matched, '^candle$', "match pattern-6 {$target} re");
$target = 'Candlestick';
ok(!$re->match($target), "match pattern-6 {$target}");
// }
$re = new Regexp_Assemble(['track' => 1]);
$re->add('^ab-(\\d+)-(\\d+)')->add('^ac-(\\d+)')->add('^nothing')->add('^ad-((\\d+)-(\\d+))');
// SKIP: {
// skip( "/?{...}/ and \\d+ cause a panic in this version of perl ($])", 15 ) if $PERL_VERSION_5_005;
ok(!$re->capture(), 'match p7 no prior capture');
ok($re->match('nothing captured'), 'match p7-1');
is(count($re->capture()), 0, 'match p7-1 no capture');
ok($re->match('ac-417 captured'), 'match p7-2');
$capture = $re->capture();
is(count($capture), 1, 'match p7-2 capture');
is($capture[0], 417, "match p7-2 value 0 ok");
ok($re->match('ab-21-17 captured'), 'match p7-3');
$capture = $re->capture();
is(count($capture), 2, 'match p7-3 capture');
is($capture[0], 21, "match p7-3 value 0 ok");
is($capture[1], 17, "match p7-3 value 1 ok");
function _do_reduce($path, $ctx)
{
// my $indent = ' ' x $ctx->{depth};
$indent = str_repeat(' ', $ctx['depth']);
// my $debug = $ctx->{debug};
$debug = $ctx['debug'];
// my $ra = Regexp::Assemble->new(chomp=>0);
$ra = new Regexp_Assemble(array('chomp' => 0));
// $ra->debug($debug);
$ra->__debug = $debug;
//$ra->__debug = 255;
// $debug and print "# $indent| do @{[_dump($path)]}\n";
if ($debug) {
echo "# {$indent}| do " . $this->_dump($path) . "\n";
}
// $ra->_insertr( $_ ) for
// # When nodes come into the picture, we have to be careful
// # about how we insert the paths into the assembly.
// # Paths with nodes first, then closest node to front
// # then shortest path. Merely because if we can control
// # order in which paths containing nodes get inserted,
// # then we can make a couple of assumptions that simplify
// # the code in _insert_node.
// sort {
// scalar(grep {ref($_) eq 'HASH'} @$a)
// <=> scalar(grep {ref($_) eq 'HASH'} @$b)
// ||
// _node_offset($b) <=> _node_offset($a)
// ||
// scalar @$a <=> scalar @$b
// }
// @$path
// ;
$self = $this;
//php5.4からは thisは暗黙なのだが・・ php5.3で動かしたいので、一度代入して自分で束縛する.
$_temp_path = $this->_perl_sort(function ($a, $b) use($self) {
$scalar_count_a = count($self->_perl_grep(function ($_) {
return is_array($_);
}, $a));
$scalar_count_b = count($self->_perl_grep(function ($_) {
return is_array($_);
}, $b));
if ($scalar_count_a > $scalar_count_b) {
return 1;
} else {
if ($scalar_count_a < $scalar_count_b) {
return -1;
}
}
$temp_b = $self->_node_offset($b);
$temp_a = $self->_node_offset($a);
if ($temp_b > $temp_a) {
return 1;
} else {
if ($temp_b < $temp_a) {
return -1;
}
}
$temp_a = count($a);
$temp_b = count($b);
if ($temp_a > $temp_b) {
return 1;
} else {
if ($temp_a < $temp_b) {
return -1;
}
}
//挙動を見ていると辞書順な気がする・・・
return strcmp(is_array($a) ? json_encode($a) : $a, is_array($b) ? json_encode($b) : $b);
// return 0;
}, $path);
foreach ($_temp_path as $_) {
$ra->_insertr($_);
}
// $path = $ra->_path;
$path = $ra->__path;
// my $common = [];
$common = array();
// push @$common, shift @$path while( ref($path->[0]) ne 'HASH' );
//これでいいんかな？
while (count($path) > 0 && !is_array($path[0])) {
// $common = $this->_perl_push2($common, array_shift($path));
$_temp_node = array_shift($path);
if (is_array($_temp_node)) {
$common = array_merge($common, $_temp_node);
} else {
$common[] = $_temp_node;
}
}
// my $tail = scalar( @$path ) > 1 ? [@$path] : $path->[0];
$tail = count($path) > 1 ? $path : $path[0];
// $debug and print "# $indent| _do_reduce common={[_dump($common)]} tail=@{[_dump($tail)]}\n";
if ($debug) {
echo "# {$indent}| _do_reduce common=" . $this->_dump($common) . " tail=" . $this->_dump($tail) . "\n";
}
// return ($common, $tail);
return array($common, $tail);
//}
}
function permute($target, $path)
{
foreach (range(0, 4) as $x1) {
foreach (range(0, 4) as $x2) {
if ($x2 == $x1) {
continue;
}
foreach (range(0, 4) as $x3) {
if (count(array_filter([$x1, $x2], function ($_) use($x3) {
return $_ == $x3;
}))) {
continue;
}
foreach (range(0, 4) as $x4) {
if (count(array_filter([$x1, $x2, $x3], function ($_) use($x4) {
return $_ == $x4;
}))) {
continue;
}
foreach (range(0, 4) as $x5) {
if (count(array_filter([$x1, $x2, $x3, $x4], function ($_) use($x5) {
return $_ == $x5;
}))) {
continue;
}
$ra = new Regexp_Assemble();
$ra->insert($path[$x1])->insert($path[$x2])->insert($path[$x3])->insert($path[$x4])->insert($path[$x5]);
is_deeply($ra->__path, $target, 'join: /' . join('/ /', array(join('', $path[$x1]), join('', $path[$x2]), join('', $path[$x3]), join('', $path[$x4]), join('', $path[$x5]))) . '/\\n' . $ra->dump() . ' versus ' . $ra->_dump($target) . "\n");
}
}
}
}
}
}
$ra = new Regexp_Assemble();
foreach ($list as $p) {
$ra->insert(str_split($p));
}
$ra->_reduce();
is_deeply($ra->__path, ['t', ['' => 0, 'h' => ['h', ['' => 0, 'o' => ['o', 'r']]], 'r' => ['r']], 'o', 'u', 'g', 'h'], join(' ', $list));
$list = ['tough', 'though', 'trough', 'through', 'thorough'];
$ra = new Regexp_Assemble();
foreach ($list as $p) {
$ra->insert(str_split($p));
}
$ra->_reduce();
is_deeply($ra->__path, ['t', ['' => 0, 'h' => ['h', ['o' => [['' => 0, 'o' => ['o', 'r']]], 'r' => ['r']]], 'r' => ['r']], 'o', 'u', 'g', 'h'], join(' ', $list));
$list = ['tit', 'titanate', 'titania', 'titanite', 'titano', 'tite', 'titi', 'titian', 'titien', 'tittie'];
$ra = new Regexp_Assemble();
foreach ($list as $p) {
$ra->insert(str_split($p));
}
$ra->_reduce();
is_deeply($ra->__path, ['t', 'i', 't', ['' => 0, 'a' => ['a', 'n', ['a' => ['a', 't', 'e'], 'i' => ['i', ['a' => ['a'], 't' => ['t', 'e']]], 'o' => ['o']]], 'i' => ['i', ['' => 0, 'a' => [['e' => ['e'], 'a' => ['a']], 'n']]], 't' => [['' => 0, 't' => ['t', 'i']], 'e']]], join(' ', $list));
$ra = new Regexp_Assemble();
$ra->add('dasin');
$ra->add('dosin');
$ra->add('dastin');
$ra->add('dostin');
$ra->_reduce();
is_deeply($ra->__path, ['d', ['a' => ['a'], 'o' => ['o']], 's', ['' => 0, 't' => ['t']], 'i', 'n'], 'dasin/dosin/dastin/dosting');
/*
is( $_, $fixed, '$_ has not been altered' );
*/
echo "===OK===\n";
<?php
require_once "Regexp_Assemble.php";
for ($i = 0; $i < 10000; ++$i) {
$reg = new Regexp_Assemble();
$reg->add('神岸あかり');
$reg->add('赤座あかり');
$reg->add('黒座あかり');
$str = $reg->re();
$reg = new Regexp_Assemble();
$reg->add('スティーブ・ジョブズ');
$reg->add('スティーブ・ウォズアニック');
$str = $reg->re();
$reg = new Regexp_Assemble();
$reg->add('お兄ちゃま');
$reg->add('あにぃ');
$reg->add('お兄様');
$reg->add('おにいたま');
$reg->add('兄上様');
$reg->add('にいさま');
$reg->add('アニキ');
$reg->add('兄くん');
$reg->add('兄君さま');
$reg->add('兄チャマ');
$reg->add('兄や');
$str = $reg->re();
}
function match_list($tag, $patt, $test)
{
$re = new Regexp_Assemble();
$re->add($patt);
$rela = new Regexp_Assemble();
$rela->lookahead(1)->add($patt);
foreach ($test as $str) {
like($str, '/^' . $re . '$/', "re {$tag}: {$str}");
like($str, '/^' . $rela . '$/', "rela {$tag}: {$str}");
}
}
$_ = rtrim($_);
if ($_ === '') {
continue;
}
$list[] = $_;
}
if (count($list) <= 0) {
$out['wordlist'] = "内容がないよう\nここに改行区切りで単語を入れよう。\n";
$out['regexp'] = '';
$out['code'] = '';
$out['bad'] = false;
$out = hh($out);
include "index.tpl";
}
//正規表現生成
$assemble = new Regexp_Assemble();
foreach ($list as $_) {
$assemble->add(_perl_quotemeta($_));
}
$out['regexp'] = $assemble->as_string();
//念のためマッチテストする
$bad = false;
foreach ($list as $_) {
if (!preg_match('/^' . $out['regexp'] . '$/u', $_)) {
$bad = true;
}
}
mylog("入力単語:" . join("\t", $list) . "\t出力正規表現:" . $out['regexp'] . "\t判定:{$bad}\r\n");
$out['code'] = 'require_once(\'Regexp_Assemble.php\'); //正規表現自動生成モジュール' . "\n";
$out['code'] .= "\n";
$out['code'] .= "\n";
$ptr = $s;
// $ptr =~ s/\\/\\\\/;
$ptr = preg_replace('/\\\\/', '/\\\\\\\\/', $ptr);
// $ptr =~ s/\n/\\n/;
$ptr = preg_replace('/\\n/', '/\\\\n/', $ptr);
// my $bug_success = ($s =~ /\n/) ? 0 : 1;
$bug_success = preg_match("/\n/", $s) ? 0 : 1;
$bug_fail = 1 - $bug_success;
////保留
// is( preg_match("/$re/" , $s ) ? $bug_success : $bug_fail, $ok,
// "Folded meta pairs behave as list for \\$meta ($ptr,ok=$ok/$bug_success/$bug_fail)"
// );
is(preg_match("/{$re_fold}/", $s) ? 1 : 0, $ok, "Unfolded meta pairs behave as list for \\{$meta} ({$ptr},ok={$ok})");
}
}
$u = new Regexp_Assemble(['unroll_plus' => 1]);
$u->add("a+b", 'ac');
$str = $u->as_string();
is($str, 'a(?:a*b|c)', 'unroll plus a+b ac');
$u->add("\\LA+B", "ac");
$str = $u->as_string();
is($str, 'a(?:a*b|c)', 'unroll plus \\LA+B ac');
$u->add("\\Ua+?b", "AC");
$str = $u->as_string();
is($str, 'A(?:A*?B|C)', 'unroll plus \\Ua+?b AC');
$u->add('\\d+d', '\\de', '\\w+?x', '\\wy');
$str = $u->as_string();
is($str, '(?:\\w(?:\\w*?x|y)|\\d(?:\\d*d|e))', 'unroll plus \\d and \\w');
//
//PHPではUTF-8以外のマルチバイトは扱えません。
// $u->add( '\\xab+f', '\\xabg', '\\xcd+?h', '\\xcdi');
function lcmp($a, $b)
{
$r = new Regexp_Assemble();
is_deeply($r->_lex($a), array($a), "_lex {$a} source line {$b}");
}
)
|m
(?:
inatur
|ers
)
|
(?:
ligh
|jec
)
t
|e
(?:
rwe
|d
)
|athwe
|press
|vil
)
ed', 'indent de.*ed');
//追加
$r = new Regexp_Assemble();
is($r->add('unimped', 'unimpeded', 'unimpelled')->as_string(), 'unimpe(?:(?:de)?|lle)d', 'unimped unimpeded unimpelled');
$r = new Regexp_Assemble();
is($r->add('tiao', 'tie', 'tien', 'tin', 'tine', 'tinea', 'tinean', 'tineine', 'tininess', 'tinnet', 'tinniness', 'tinosa', 'tinstone', 'tint', 'tinta', 'tintie', 'tintiness', 'tintist', 'tisane', 'tit', 'titanate', 'titania', 'titanite', 'titano', 'tite', 'titi', 'titian', 'titien', 'tittie')->as_string(), 'ti(?:n(?:t(?:i(?:ness|st|e)|a)?|e(?:an?|ine)?|n(?:iness|et)|iness|stone|osa)?|t(?:an(?:i(?:te|a)|ate|o)|i(?:[ae]n)?|(?:ti)?e)?|sane|en?|ao)', 'tiao tie .... titien tittie');
/*
is( $_, $fixed, '$_ has not been altered' );
*/
echo "===OK===\n";