XPath技术咨询：如何筛选首个指定class的div下不含关键词的文本节点

Let's break down what's going wrong with your current expression and build the correct one for your use case.

What's Wrong With Your Current XPath? #

Your expression (//div[@class='blabla'])[1][not(contains(text(), 'bananas'))] has two key issues:

1. text() on the parent div only targets text nodes directly inside the div itself—not the text inside its child p tags. Since your "bananas" text lives in a nested p, this condition never actually checks for the keyword you care about.
2. You're filtering the entire parent div instead of the individual child nodes you want to exclude. Even if the condition worked, it would discard the entire first blabla div if any child had "bananas", which isn't what you want (you just want to skip the problematic child node, not the whole container).

The Correct XPath for Your Example #

To get only the text from p tags inside the first blabla div, excluding any p that contains "bananas", use this expression:

(//div[@class='blabla'])[1]/p[not(contains(., 'bananas'))]/text()

Let's Break It Down: #

• (//div[@class='blabla'])[1]: Pinpoints the first div with class blabla (avoids matching the nested inner blabla div that's a child of the first one).
• /p: Targets only direct child p tags of that first div (if you wanted all descendant p tags, use //p instead—but your example doesn't include the inner div's text, so direct children align with your desired result).
• [not(contains(., 'bananas'))]: Filters out any p tag where the node's content (. refers to the current p node's full text) includes the word "bananas".
• /text(): Extracts the raw text content from the matching p tags.

If You Need All Descendant Text (Including Nested Nodes, Excluding Keywords) #

If you ever want to capture all text inside the first blabla div (not just direct child ps) but still exclude any node that contains "bananas", use this:

(//div[@class='blabla'])[1]//*[not(contains(., 'bananas'))]/text()

This will grab text from any descendant element, as long as that element doesn't contain the excluded keyword.

Testing With Your Sample HTML #

For your given markup:

<div class="blabla"> 
  <p>I like bananas</p> 
  <p>I also like apples</p> 
  <div class="blabla"> 
    <p>some text</p> 
  </div>
</div>

The first XPath will return exactly:

I also like apples

Which matches your desired result.

内容的提问来源于stack exchange，提问作者Adrien