在preg_match请求中获取所有匹配项

Question

I'm having the following problem, i have that structure:

$table = '
<table>
    <tbody>
        <tr valign="top">
            <td>foo</td>
            <td>bar</td>
        </tr>
    </tbody>
</table>
<table>
    <tbody>
        <tr valign="top">
            <td>bee</td>
            <td>dog</td>
        </tr>
    </tbody>
</table>';

I'm trying to retrieve an array with all <tr> but with no success. The closest pattern I've could made it, return all messed up.

$pattern = "/<tr valign[^>]*>(.*)<\/tr>/s";
preg_match_all($pattern, $table, $matches, PREG_PATTERN_ORDER);

If i put var_dump($matches), I want an array that returns:

array(
    [0] => "<td>foo</td><td>bar</td>",
    [1] => "<td>bee</td><td>dog</td>"
);

...or something close to that.

But I receive:

string(301) "
    foo
    bar
    "
<table>
        <tbody>
            <tr valign="top">
                <td>bee</td>
                <td>dog</td>
            </tr>
    </tbody></table>

Anyone know what I'm doing wrong?

Thanks in advance.

Answer 1

You must make your quantifier lazy: .* => .*?

When you use a greedy quantifier, .* will take all possible characters, When you use a lazy quantifier, .*? will take the minimum number of characters.

When you use a lazy quantifier, the regex engine will take characters one by one and test the pattern completion for each character.

When you use a greedy quantifier (default behavior) the regex engine will take all possible characters (until the end in your case) and will backtrack character by character until the pattern completion succeed.

Notes:

It is useless to add PREG_PATTERN_ORDER since it is the default set of preg_match_all.

DOMDocument is probably a more adapted tool to deal with html. Example:

$dom = new DOMDocument();
@$dom->loadHTML($table);

$trs = $dom->getElementsByTagName('tr');

$results = array();

foreach ($trs as $tr) {
    if ($tr->hasAttribute('valign')) {
        $children = $tr->childNodes;

        $tmp = '';
        foreach ($children as $child) {
            $tmp .= trim($dom->saveHTML($child));
        }
        if (!empty($tmp)) $results[] = $tmp;
    }
}

echo htmlspecialchars(print_r($results, true));