PHP中移除字符串内除末尾来源链接外的所有锚标签

要实现这个需求，我们可以分两步走：先把末尾的来源链接部分单独提取出来，处理剩余文本移除所有锚标签后，再把来源链接拼接回去。这种方法逻辑清晰，不容易出错，也避免了复杂正则的坑。

方案代码示例： #

$str = "It’s been a rough couple of days for <a href='aaa.com'>Apple</a>, as the company is still trying to fix a massive glitch in Facetime that allows consumers to hear audio from users added to group calls before they pick up. It’s not known how long the glitch has existed, but the details of it became widespread knowledge yesterday and prompted Apple to disable the feature entirely. As it turns out, a 14-year-old Arizona teenager discovered the glitch last week while setting up a group chat of Fortnite players, inadvertently finding himself at ground zero for a massive privacy breach debacle. The teenager discovered the issue on January 20th while setting up a group chat for his friends ahead of a weekend Fortnite session. As he added his pals into the group video chat, he realized that he could hear audio from users he had just added to the group chat before they’d even picked up. He notified his mom, Michele Thompson, who then tried to inform Apple about the glitch. The <a href='bbb.com'>proactive</a> mother went through several mediums in an effort to garner a response from Apple, including phone, fax, Facebook, and Twitter, where she even tried to reach out to Apple CEO Tim Cook. Unfortunately, none of these mediums garnered an immediate response from Apple. Source: <a href='sou.com'>Source</a>";

// 第一步：提取末尾的来源链接部分
preg_match('/Source: <a[^>]*>Source<\/a>/', $str, $matches);
$sourceLink = isset($matches[0]) ? $matches[0] : '';

// 第二步：移除原字符串中的来源链接部分，处理剩余文本删除所有锚标签
$contentWithoutSource = str_replace($sourceLink, '', $str);
$cleanedContent = preg_replace('/<\/?a[^>]*>/', '', $contentWithoutSource);

// 第三步：拼接处理后的内容和保留的来源链接
$result = trim($cleanedContent) . ' ' . $sourceLink;

echo $result;

代码解释： #

1. 提取来源链接：用preg_match精准匹配末尾的Source: <a...>Source</a>部分，确保我们只保留这一段。
2. 清理剩余内容：先把原字符串中的来源链接去掉，再用你原来的正则移除所有锚标签，得到干净的正文内容。
3. 拼接结果：把清理后的正文和保留的来源链接拼接起来，就得到了想要的结果。

另一种方案：使用正则否定前瞻（进阶） #

如果你想用单一正则完成，可以利用否定前瞻来排除末尾的来源锚标签。不过要注意匹配的精准性：

$result = preg_replace('/<\/?a[^>]*>(?!Source<\/a>)/', '', $str);

这个正则会匹配所有锚标签，除了那些后面跟着Source</a>的（也就是最后那个来源链接的标签）。不过这种方法依赖于来源链接的固定文本（Source），如果来源文本可能变化，还是第一种方法更稳妥。

内容的提问来源于stack exchange，提问作者Thiyagarajan