如何模仿StackOverflow自动链接行为
用PHP怎样才能模仿StackOverflow的自动链接行为（BTW非常酷）？
例如，以下URL：
http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior
被转换成这个：
<a title="how to mimic stackoverflow auto link behavior" rel="nofollow" href="http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior">stackoverflow.com/questions/1925455/…</a> 在这种情况下，我并不关心title属性。
和这个：
http://pt.php.net/manual/en/function.base-convert.php#52450
被转换成这个：
<a rel="nofollow" href="http://pt.php.net/manual/en/function.base-convert.php#52450">pt.php.net/manual/en/…</a>
我怎样才能在PHP中做出类似的function？
PS：请查看我对这个问题的评论以获得更多的例子和行为。
试试这个。 URL匹配正则expression式模式来自Daring Fireball 。
/** * Replace links in text with html links * * @param string $text * @return string */ function auto_link_text($text) { // a more readably-formatted version of the pattern is on http://daringfireball.net/2010/07/improved_regex_for_matching_urls $pattern = '(?i)\b((?:[az][\w-]+:(?:/{1,3}|[a-z0-9%])|www\d{0,3}[.]|[a-z0-9.\-]+[.][az]{2,4}/)(?:[^\s()<>]+|\(([^\s()<>]+|(\([^\s()<>]+\)))*\))+(?:\(([^\s()<>]+|(\([^\s()<>]+\)))*\)|[^\s`!()\[\]{};:\'".,<>?«»“”'']))'; $callback = create_function('$matches', ' $url = array_shift($matches); $url_parts = parse_url($url); $text = parse_url($url, PHP_URL_HOST) . parse_url($url, PHP_URL_PATH); $text = preg_replace("/^www./", "", $text); $last = -(strlen(strrchr($text, "/"))) + 1; if ($last < 0) { $text = substr($text, 0, $last) . "…"; } return sprintf(\'<a rel="nofollow" href="%s">%s</a>\', $url, $text); '); return preg_replace_callback($pattern, $callback, $text); }
input文字：
This is my text. I wonder if you know about asking questions on StackOverflow: Check This out http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior Also, base_convert php function? http://pt.php.net/manual/en/function.base-convert.php#52450 http://pt.php.net/manual/en/function.base-convert.php?wtf=hehe#52450
输出文本：
This is my text. I wonder if you know about asking questions on StackOverflow: Check This out <a rel="nofollow" href="http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior">stackoverflow.com/questions/1925455/…</a> Also, base_convert php function? <a rel="nofollow" href="http://pt.php.net/manual/en/function.base-convert.php#52450">pt.php.net/manual/en/…</a> <a rel="nofollow" href="http://pt.php.net/manual/en/function.base-convert.php?wtf=hehe#52450">pt.php.net/manual/en/…</a>
这是基于相同的daringfireball.net正则expression式 ，但增加了一些比Eric Coleman的例子更多的逻辑，以及最大URL深度configuration（SO似乎是50），URL截断时的最大path深度（SO似乎是2）和省略号（ … ）。
据我所知，这个复制了所有的URL重写function，至less到目前为止在这里的评论和回应中所讨论的内容。
function auto_link_text($text) { $pattern = '#\b(([\w-]+://?|www[.])[^\s()<>]+(?:\([\w\d]+\)|([^[:punct:]\s]|/)))#'; return preg_replace_callback($pattern, 'auto_link_text_callback', $text); } function auto_link_text_callback($matches) { $max_url_length = 50; $max_depth_if_over_length = 2; $ellipsis = '…'; $url_full = $matches[0]; $url_short = ''; if (strlen($url_full) > $max_url_length) { $parts = parse_url($url_full); $url_short = $parts['scheme'] . '://' . preg_replace('/^www\./', '', $parts['host']) . '/'; $path_components = explode('/', trim($parts['path'], '/')); foreach ($path_components as $dir) { $url_string_components[] = $dir . '/'; } if (!empty($parts['query'])) { $url_string_components[] = '?' . $parts['query']; } if (!empty($parts['fragment'])) { $url_string_components[] = '#' . $parts['fragment']; } for ($k = 0; $k < count($url_string_components); $k++) { $curr_component = $url_string_components[$k]; if ($k >= $max_depth_if_over_length || strlen($url_short) + strlen($curr_component) > $max_url_length) { if ($k == 0 && strlen($url_short) < $max_url_length) { // Always show a portion of first directory $url_short .= substr($curr_component, 0, $max_url_length - strlen($url_short)); } $url_short .= $ellipsis; break; } $url_short .= $curr_component; } } else { $url_short = $url_full; } return "<a rel=\"nofollow\" href=\"$url_full\">$url_short</a>"; }
示例input：
This is my text. I wonder if you know about asking questions on StackOverflow: Check This out http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior Also, base_convert php function? http://pt.php.net/manual/en/function.base-convert.php#52450 http://pt.php.net/manual/en/function.base-convert.php?wtf=hehe#52450 http://ab/c/d/e/f/test and http://ab/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/z/y/w/z/test
示例输出：
This is my text. I wonder if you know about asking questions on StackOverflow: Check This out <a rel="nofollow" href="http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior">http://stackoverflow.com/questions/1925455/…</a> Also, base_convert php function? <a rel="nofollow" href="http://pt.php.net/manual/en/function.base-convert.php#52450">http://pt.php.net/manual/en/…</a> <a rel="nofollow" href="http://pt.php.net/manual/en/function.base-convert.php?wtf=hehe#52450">http://pt.php.net/manual/en/…</a> <a rel="nofollow" href="http://ab/c/d/e/f/test">http://ab/c/d/e/f/test</a> and <a rel="nofollow" href="http://ab/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/z/y/w/z/test">http://ab/c/d/…</a>
这将把样本string转换成你以后的样子。 由于来自不同的来源，而不是单独的URL，所以我忽略了title ，并且您说这不重要。
<?php $urlInput="http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior"; preg_match('@http://(?:www\.)?(\S+/)\S*(?:\s|$)@i', $urlInput, $matches); print('<a rel="nofollow" href="' . trim($matches[0]) . '">' . $matches[1] . '...</a>'); ?>
根据需要进行扩展以扫描文本。
如果您只想匹配一定数量的URLpath元素，请使用以下RE：
'@http://(?:www\.)?((?:\S+?/){1,3})\S*(?:\s|$)@i'
这将最多提取3个path元素（主机和最多两个目录）。 您可以改变{1,3}中的上限来定义所需的最大path元素数量。
更改结束\S允许零匹配。
如果你有一个像SO这样的可预测的URL，那么应该很容易用正则expression式来获取链接，并过滤出与模式匹配的链接。 因此，如果您的url是http://example.com/stuff/1234那么使用正则expression式findhttp://example.com/stuff/1234/how-to-mimic将会非常简单。
<?php preg_match('/http:\/\/example.com\/(\w*)\/(\d)[\/*]/', $text, $matches); if (is_array($matches)) { foreach ($matches as $match) { // do something... } } ?>
根据Kevin Brock的回答，但允许configuration参数（文件夹深度和URL长度），并接受没有尾部斜线的URL：
$url = 'http://www.stackoverflow.com/questions/1925455/how-to-mimic-stackoverflow-auto-link-behavior'; $output = ''; $params = array ( 'length' => 10, 'depth' => 2, ); preg_match ('@http://(?:www\.)?([^/?# ]+)(/\S+)?(?=\s|$)@i', $url, $matches); if (isset ($matches[2])) { $parts = explode('/', substr($matches[2], 1)); if (count($parts) > $params['depth'] && strlen($matches[1].$matches[2]) > $params['length']) $output = $matches[1].'/'.implode('/', array_slice($parts, 0, 2)).'/...'; else $output = $matches[1].$matches[2]; } else $output = $matches[1]; echo '<a href="'.$matches[0].'">'.$output.'</a>';
希望这可以帮助
请参阅正则expression式（正则expression式）以匹配URL ：
https?://([-\w\.]+)+(:\d+)?(/([\w/_\.]*(\?\S+)?)?)?PHP示例：自动链接URL的内部文本。
$text = preg_replace('@(https?://([-\w\.]+)+(:\d+)?(/([\w/_\.]*(\?\S+)?)?)?)@', '<a href="$1">$1</a>', $text);
