PHP中处理字符串替换冲突：利用preg_replace实现优先级匹配

$things = ["apple", "apple pie", "baked apple"];
$toReplace = "Henry ate an apple then a whole apple pie and a baked apple, too.";
$output = $toReplace;

foreach($things as $thing) {
    $output = str_replace($thing, "<i>".$thing."</i>", $output);
}
echo $output;
// 可能会输出：Henry ate an <i>apple</i> then a whole <i><i>apple</i> pie</i> and a <i>baked <i>apple</i></i>, too.
// 或者：Henry ate an <i>apple</i> then a whole <i>apple</i> pie and a baked <i>apple</i>, too.

$things_unsorted = ['apple', 'apple pie', 'baked apple']; // 未排序的数组
$toReplace = 'Henry ate an apple then a whole apple pie and a baked apple, too.';

// 错误示例：如果'apple'在'apple pie'之前，'apple'会被优先匹配
// 导致 'apple pie' 中的 'apple' 部分被替换
// echo preg_replace('/' . implode('|', $things_unsorted) . '/i', '<i>$0</i>', $toReplace);
// 结果可能仍然是：Henry ate an <i>apple</i> then a whole <i><i>apple</i> pie</i> and a <i>baked <i>apple</i></i>, too.

<?php

/**
 * 替换字符串数组中的项，避免子字符串冲突和嵌套标签。
 *
 * @param array  $things    要替换的字符串数组。
 * @param string $toReplace 待处理的原始字符串。
 * @param string $tag       用于包裹替换项的HTML标签（例如 'i'）。
 * @return string 处理后的字符串。
 */
function replaceConflictingStrings(array $things, string $toReplace, string $tag = 'i'): string
{
    // 1. 按照字符串长度从长到短排序，确保较长的字符串优先匹配
    usort($things, function($a, $b) {
        return strlen($b) - strlen($a);
    });

    // 2. 对每个替换项进行正则表达式转义，防止特殊字符被误解释
    $things_escaped = array_map('preg_quote', $things, array_fill(0, count($things), '/')); // 确保转义斜杠

    // 3. 将所有转义后的替换项用 '|' 连接，构建正则表达式模式
    // '/i' 修饰符表示不区分大小写匹配
    $pattern = '/' . implode('|', $things_escaped) . '/i';

    // 4. 使用 preg_replace 进行替换，$0 代表整个匹配到的字符串
    return preg_replace($pattern, "<{$tag}>$0</{$tag}>", $toReplace);
}

// 示例用法
$things_to_replace = ["apple", "apple pie", "baked apple", "pie"]; // 包含冲突的字符串
$main_string = "Henry ate an apple then a whole apple pie and a baked apple, too. I love pie!";

echo "原始字符串: " . $main_string . PHP_EOL;

$result = replaceConflictingStrings($things_to_replace, $main_string, 'b'); // 使用<b>标签
echo "处理后字符串: " . $result . PHP_EOL;

// 预期输出: Henry ate an <b>apple</b> then a whole <b>apple pie</b> and a <b>baked apple</b>, too. I love <b>pie</b>!

// 另一个示例
$things_to_replace_2 = ["PHP", "PHP教程", "教程"];
$main_string_2 = "这是一个PHP教程，学习PHP非常有用。";
echo "\n原始字符串2: " . $main_string_2 . PHP_EOL;
$result_2 = replaceConflictingStrings($things_to_replace_2, $main_string_2, 'strong');
echo "处理后字符串2: " . $result_2 . PHP_EOL;
// 预期输出: 这是一个<strong>PHP教程</strong>，学习<strong>PHP</strong>非常有用。
?>