求匹配不含@href的<test>元素的正则表达式及问题解决

Hey there! Let's sort out that regex problem you're facing. The issue with your original pattern <test\s+abc.[^href]*>.* is that [^href] is a character set, not a string exclusion. It tells the regex to exclude any single character that's h, r, e, or f—so when your abc attribute value contains any of those letters (like the h in abc_CHmnoph), the regex stops matching early, which is why your third <test> element wasn't picked up.

The Correct Regex #

Here's a pattern that will reliably match all <test> elements without an href attribute:

<test\s+(?:(?!href=)[^>])+>.*?</test>

Breakdown of the Pattern #

Let's break down what each part does to make it clear:

• <test\s+: Matches the start of the <test> tag, followed by one or more whitespace characters (to account for attributes).
• (?:(?!href=)[^>])+: This is the key fix. The (?!href=) is a negative lookahead—it checks that the substring href= does NOT start at the current position. Then [^>] matches any character that isn't a closing >, and the + repeats this logic for the entire attribute section of the tag. This ensures we only match <test> tags that never contain href=.
• >.*?</test>: Matches the closing > of the opening tag, then uses non-greedy .*? to capture all content inside the tag until we hit the closing </test> (this prevents over-matching if multiple tags are present).

Testing with Your Sample Content #

Using your test input:

<test abc="xyz_CHCFGRc/abc_CHmnop" href="sdddzus.xml">text</test> <test abc="abc_abc">text23</test> <test abc="123_ABCc/abc_CHmnoph">text42</test>

This regex will correctly match exactly the two elements you want:

1. <test abc="abc_abc">text23</test>
2. <test abc="123_ABCc/abc_CHmnoph">text42</test>

Quick Note on Regex vs. HTML Parsers #

While this regex works for your specific use case, it's worth mentioning that regex isn't ideal for parsing HTML in general (edge cases like nested tags or attribute values with > can break it). If you're working with more complex HTML, using a dedicated HTML parser would be more robust—but for this simple scenario, the regex above does the trick!

内容的提问来源于stack exchange，提问作者Shil