我提供了以下HTML。类名始终相同。仅标签之间的文本不同,并且长度和内容也不同。
<a>
<span class="xxx">Not this text <span class="yyy">not this text</span> <span class="zzz">This is</span> the required text <q class="aaa">this not</q></span>
</a>
我如何在类“ zzz”的标签和行尾之间提取内容,但是结果中应该not包括类“ aaa”的元素?有可能吗?
类别为“ aaa”的元素可能存在或不存在:
<a>
<span class="xxx">Not this text <span class="yyy">not this text</span> <span class="zzz">This is</span> the required text</span>
</a>
预期结果应该是:
This is the required text
还可能存在“所需文本”部分:
<a>
<span class="xxx">Not this text <span class="yyy">not this text</span> <span class="zzz">This is</span></span>
</a>
所以结果应该是:
This is
我使用DOMXPath在PHP中尝试此方法。
我不一定不知道如何使用XPath来做到这一点,但是这是一种无需XPath就可以做到的方式。
function walk(DOMNode $node, $skipParent = false) {
if (!$skipParent) {
yield $node;
}
if ($node->hasChildNodes()) {
foreach ($node->childNodes as $n) {
yield from walk($n);
}
}
}
$html = <<<'HTML'
<span class="xxx">
Not this text
<span class="yyy">not this text</span>
<span class="zzz">This is</span>
the required text
<q class="aaa">this not</q>
</span>
HTML;
$dom = new DOMDocument;
$dom->loadHTML($html, LIBXML_HTML_NOIMPLIED | LIBXML_HTML_NODEFDTD);
$count = 0;
foreach(walk($dom->firstChild) as $node) {
if (!($node instanceof DOMText) && $node->hasAttribute('class') && $node->getAttribute('class') === 'xxx') {
foreach(walk($node) as $n) {
if (isset($content)) {
$count++;
}
if (!($n instanceof DOMText) && $n->hasAttribute('class') && $n->getAttribute('class') === 'zzz') {
$content = $n->textContent;
}
if (isset($content) && $n instanceof DOMText && $count == 2) {
$content .= " " . $n->textContent;
break 2;
}
}
}
}
var_dump($content);
无论是否存在"the required text"
部分,这都会为您提供所需的结果。