需從任意 HTML 頁面提取所有表單結構 (action、method、input/select/textarea 的 name/type/value/options) 以建立自動化測試、爬蟲或安全掃描器。本文提供完整正則解析器實作，逐層拆解 form → 元素 → 屬性。

目標 HTML 範例 (含兩個 form、多種元素類型)

<form action="register.php" method="post">
  <input type="hidden" name="cookie" value="SJKjki80KJ8jkl2" />
  <input type="text" name="name" required />
  <input name="password" type="password" required="required" />
  <input name="age" value="22" required />
  <input type="radio" name="sex" value="1" /> 男
  <input type="radio" name="sex" value="0" /> 女
  <select name="area">
    <option value="jiangsu">江蘇</option>
    <option value="shandong">山東</option>
  </select>
  <textarea name="summary">個人簡介</textarea>
  <input type="submit" name="register" value="註冊" />
</form>

<form action="login.php" method="post">
  <input type="text" name="name" required />
  <input name="password" type="password" />
  <textarea name="remark">2</textarea>
  <input type="submit" name="register" value="登入" />
</form>

解析器核心架構：分層正則 + 陣列組裝

<?php
function parse_forms($html) {
    $forms = [];
    // 1. 擷取所有 
...
 區塊 (s 單行模式讓 . 匹配換行)
    if (!preg_match_all('/(.*?)<\/form>/si', $html, $m)) return $forms;

    for ($i = 0; $i < count($m[0]); $i++) {
        $formAttrs = $m[1][$i]; // form 標籤屬性字串
        $formInner = $m[2][$i]; // form 內部 HTML

        $forms[$i]['action'] = attr($formAttrs, 'action');
        $forms[$i]['method'] = attr($formAttrs, 'method') ?: 'get';

        // 2. 解析 input
        if (preg_match_all('/]*>/si', $formInner, $im)) {
            foreach ($im[0] as $inp) {
                $forms[$i]['inputs'][] = [
                    'name'  => attr($inp, 'name'),
                    'type'  => attr($inp, 'type') ?: 'text',
                    'value' => attr($inp, 'value'),
                ];
            }
        }

        // 3. 解析 textarea
        if (preg_match_all('/]*>(.*?)<\/textarea>/si', $formInner, $tm)) {
            foreach ($tm[0] as $idx => $tag) {
                $forms[$i]['textarea'][] = [
                    'name'  => attr($tag, 'name'),
                    'value' => $tm[1][$idx], // 內容在捕獲組 1
                ];
            }
        }

        // 4. 解析 select + options
        if (preg_match_all('/(.*?)<\/select>/si', $formInner, $sm)) {
            foreach ($sm[0] as $idx => $tag) {
                $select = ['name' => attr($sm[1][$idx], 'name'), 'options' => []];
                // 解析 option
                if (preg_match_all('/]*value=["\']([^"\']*)["\'][^>]*>/si', $sm[2][$idx], $om)) {
                    $select['options'] = $om[1];
                }
                $forms[$i]['select'][] = $select;
            }
        }
    }
    return $forms;
}

// 通用屬性提取輔助函數
function attr($tag, $name) {
    if (preg_match('/' . $name . '\s*=\s*(["\'])(.*?)\1/i', $tag, $m)) {
        return $m[2];
    }
    return null;
}

// 使用範例
$html = file_get_contents('1.html');
$forms = parse_forms($html);
print_r($forms);

輸出結構範例

Array
(
    [0] => Array
        (
            [action] => register.php
            [method] => post
            [inputs] => Array
                (
                    [0] => Array ( [name] => cookie [type] => hidden [value] => SJKjki80KJ8jkl2 )
                    [1] => Array ( [name] => name [type] => text [value] => )
                    ...
                )
            [textarea] => Array ( [0] => Array ( [name] => summary [value] => 個人簡介 ) )
            [select] => Array
                (
                    [0] => Array ( [name] => area [options] => Array ( [0] => jiangsu [1] => shandong ) )
                )
        )
    [1] => Array ( ... login form ... )
)

正則局限與進階建議

• HTML 非正則語言：巢狀標籤、屬性含 `>`、註解、CDATA、動態屬性 (Vue/React 模板) 皆會導致正則失效或誤匹配。
• 生產環境請用 DOMDocument / DOMXPath：

$dom = new DOMDocument();
@$dom->loadHTML($html); // 抑制 HTML5 標籤警告
$xpath = new DOMXPath($dom);
$forms = $xpath->query('//form');
foreach ($forms as $form) {
    $action = $form->getAttribute('action');
    $inputs = $xpath->query('.//input', $form);
    // ... 結構化提取，健壯度遠超正則
}