正则表达式捕获组匹配逻辑疑问：结果为何为FILE_NAME.EXTENSION

Great question—let's break down this capture group step by step using your example string to make the logic crystal clear.

First, let's define each piece of the capture group: #

• [^.]: This is a negated character class. Inside square brackets, the ^ flips the matching rule: instead of matching specific characters, it matches any single character except a dot (.).
• +: This quantifier means "match the preceding element one or more times." So [^.]+ translates to "grab one or more characters that are definitely not dots."
• \.: An escaped dot, which matches a literal dot character (since an unescaped . in regex matches almost any character except a newline).

Applying this to your example string #

After the ^\d+_ part matches the leading numbers and underscore, your remaining string is: FILE_NAME.EXTENSION.OTHEREXTENSION

Here's how the capture group processes this:

1. The first [^.]+ starts at the F and matches every character until it hits the first dot. So it grabs FILE_NAME—it stops right at the . because [^.] can't match dots.
2. The \. then matches that first literal dot between FILE_NAME and EXTENSION.
3. The second [^.]+ starts at E and matches every character until it hits the next dot. This grabs EXTENSION—it stops at the . before OTHEREXTENSION for the same reason: [^.] won't match a dot, so it can't go past that point.

Putting it all together, the capture group ([^.]+\.[^.]+) lands exactly on FILE_NAME.EXTENSION—it can't include the .OTHEREXTENSION part because the second [^.]+ hits a dot and stops immediately.

A quick note on greediness #

You might think greediness would make the regex match more, but that's not the case here. The [^.] rule is a hard exclusion: those segments cannot include dots, so the regex can't skip over dots to grab more characters. It's a strict "stop at the first dot" rule for each [^.]+ segment.

内容的提问来源于stack exchange，提问作者Dog