匹配句子整词的正则表达式异常求助：USA-被意外匹配

Hey there! Let's dig into why your regex .*\b(?i)USA\b.* is matching USA- when you didn't expect it to.

The Root Cause #

The \b (word boundary) in regex works by matching positions between a word character (letters, digits, underscores: A-Za-z0-9_) and a non-word character (everything else, like -, !, spaces). In USA-, the A (word character) sits right next to - (non-word character), so \b sees this as a valid word boundary. That's why your regex ends up matching the whole string.

Fixes to Try #

Depending on exactly how you define a "whole word," here are a few solutions:

Option 1: Match only when USA is surrounded by spaces or string boundaries

If you want USA to be a completely standalone term (no adjacent punctuation or symbols), use this regex:

.*(?:^|\s)(?i)USA(?:\s|$).*

• (?:^|\s): Matches either the start of the string or a whitespace character (space, tab, etc.)
• (?i)USA: Case-insensitive match for USA
• (?:\s|$): Matches either a whitespace character or the end of the string

This will exclude USA- entirely, while still matching your expected cases like USA, The UsA, and USA is.

Option 2: Allow punctuation but block - and word characters

If you want to permit punctuation after USA (like USA. or USA,) but still block USA- and USAGE, use this adjusted regex with negative lookarounds:

.*(?<![^\W_])(?i)USA(?![^\W_-]).*

• (?<![^\W_]): Ensures USA isn't preceded by a letter, digit, or underscore
• (?![^\W_-]): Ensures USA isn't followed by a letter, digit, underscore, or -

This keeps your valid matches intact while excluding both USAGE and USA-.

Option 3: Quick fix to just block - after USA

If you only need to stop matching cases where USA is immediately followed by -, you can add a simple negative lookahead to your original regex:

.*\b(?i)USA(?!-)\b.*

The (?!-) checks that the character right after USA isn't -, which fixes the USA- issue without changing how other valid cases are handled.

内容的提问来源于stack exchange，提问作者MishraJi