如何用正则表达式捕获所有句号前的单个字母？

Hey there, let's work through this regex issue together!

Your original regex /^([^.]+)/ falls short for two key reasons: first, the ^ anchor ties it strictly to the start of the string, so it can't pick up later matches like the d in your example. Second, [^.]+ matches any sequence of characters until a period—not just single letters, which isn't what you need.

The Correct Regex #

Use this global regex to reliably capture every standalone single letter immediately followed by a period:

/(?<!\S)([a-zA-Z])\./g

Breakdown of Each Component: #

• (?<!\S): A negative lookbehind assertion that ensures the letter isn't preceded by a non-whitespace character. This guarantees we only target single, isolated letters (not part of longer words like test. where we'd accidentally capture the t).
• ([a-zA-Z]): The capture group that grabs the single uppercase or lowercase letter we care about.
• \.: Matches the literal period right after the letter.
• g: The global flag, which tells the regex engine to scan the entire string for all matches instead of stopping at the first one.

Example Usage (JavaScript) #

Here's how you'd extract all the captured letters in practice:

const inputStr = "a. Lorem ipsum dolor sit amet, consectetur adipiscing elit, d. Donec euismod magna velit, ac tincidunt nisl faucibus eu";
const regex = /(?<!\S)([a-zA-Z])\./g;
const capturedLetters = [];
let matchResult;

while ((matchResult = regex.exec(inputStr)) !== null) {
  capturedLetters.push(matchResult[1]);
}

console.log(capturedLetters); // Output: ["a", "d"]

Quick Edge Case Note #

This regex works seamlessly for letters at the start of the string (like your a. example) because the start of the string counts as having no non-whitespace character before it—so the lookbehind doesn't block the match.

内容的提问来源于stack exchange，提问作者Lost in google