正则表达式技术问询：匹配指定关键词且不含特定修饰词的目标行

Got it, let's tweak that regex so it only picks up Case 2 and excludes the other two. Here's the breakdown and solution:

The Core Problem #

Your current regex matches any line starting with expect or project that includes the target EPS phrase—but we need to:

1. Only target lines starting with project/Projected (exclude expect entirely)
2. Exclude lines that contain non-GAAP or adjusted (since those are present in Cases 1 and 3)

The Updated Regex #

(?i)^(?:project)(?!.*(non-GAAP|adjusted)).*(diluted\sEPS\sfrom\scontinuing\soperations)$

Let's Break It Down #

• (?i): Optional but handy—turns on case-insensitive matching, so it works for both Projected and project
• ^: Anchors the match to the start of the line
• (?:project): Matches the required starting word (non-capturing group, so it doesn't clutter your match results)
• (?!.*(non-GAAP|adjusted)): Negative lookahead—this ensures the line does NOT contain either non-GAAP or adjusted anywhere in it
• .*: Matches any characters in between (use .*? instead if you want non-greedy matching, though it's not strictly necessary here)
• (diluted\sEPS\sfrom\scontinuing\soperations): Your target EPS phrase, exactly as you need it
• $: Anchors the match to the end of the line, ensuring we're matching full lines

Why This Works for Your Cases #

• Case 1: Starts with expect—doesn't match the project opening, so it's excluded
• Case 2: Starts with Projected, has no non-GAAP/adjusted, and includes the EPS phrase—gets matched
• Case 3: Contains adjusted—the negative lookahead blocks the match

Quick Note #

If you're using a tool that doesn't support (?i) for case insensitivity, you can replace (?:project) with (?:Project|project) instead.

内容的提问来源于stack exchange，提问作者CHANDRA SUDHEER K