如何使用POSIX扩展正则表达式匹配方括号外的‘organization’单词

Got it, let's work through adapting your regex to POSIX Extended Regular Expressions (ERE)!

First, let's break down why your original regex (?:^|\s)(organization)(?!]) won't work here: POSIX ERE doesn't support non-capturing groups ((?:...)—all parentheses in ERE create capturing groups) or negative lookahead assertions ((?!])). So we need to rethink the approach to fit POSIX ERE's capabilities.

Solution 1: Accurate match for non-nested brackets (uses capture groups) #

If your application supports extracting capture group results, this is the most reliable method—especially if you might have organization inside brackets with surrounding text (like [abc organization def]). The trick is to first match and skip all bracket-enclosed content, then capture the valid organization words outside brackets:

\[.*\]|([[:<:]]organization[[:>:]])

Here's what each part does:

• \[.*\]: Matches any pair of brackets and all content inside them (the greedy .* ensures it grabs up to the closing ], which works for non-nested brackets)
• |: Acts as a logical OR—either match the bracket content, or match the capture group
• ([[:<:]]organization[[:>:]]): Captures standalone organization words outside brackets. [[:<:]] and [[:>:]] are POSIX-standard word boundaries, so we don't accidentally match parts of longer words (like organizationxyz).

When you run this regex, only the valid organization instances will show up in capture group 1—bracket-enclosed content gets matched but not captured.

Solution 2: Simplified match for specific bracket formats (no capture groups needed) #

If your use case only has brackets wrapping organization directly or with adjacent text (like [organization] or [organization name], no spaces around organization inside brackets), you can use this simpler pattern that matches valid instances directly:

(^|\s)organization([^\]]|$)

• (^|\s): Ensures we're matching the start of the string or a space before organization, so we only target standalone words
• organization: The word we want to match
• ([^\]]|$): Matches any character that isn't a closing bracket, or the end of the string—this excludes any organization that's inside brackets (since those would be followed by either ] or text leading to ]).

Testing this on your example string:

The organization “[organization name]” must contain at least one user per organization . The ID [id] for [organization] must contain digits only

It will correctly match the first organization (after "The ") and the third organization (after "per "), while ignoring the ones inside brackets.

内容的提问来源于stack exchange，提问作者user1298431