Hey there! The issue with using just \d+ is that it matches every sequence of digits in your string—including all the ones in the timestamp, the 029 after your target, and even the final 1. Let's fix that by making your regex more specific to the position or context of the number you want:

Option 1: Target the 3rd whitespace-separated field #

Your string is split into fields by spaces, and your target number is the 3rd one. You can write a regex that matches the first two fields, then captures the third:

^\S+\s+\S+\s+(\d+)

Breakdown:

• ^: Anchors the match to the start of the string (so we don't accidentally match later fields)
• \S+: Matches the first field (the date 022/03/17)
• \s+: Matches one or more spaces between fields
• \S+: Matches the second field (the timestamp 05:53:40.376949)
• \s+: Another set of spaces
• (\d+): Captures your target number 1245680

Option 2: Match based on surrounding context #

If the structure of the timestamp is consistent (date like XX/XX/XX, time like XX:XX:XX.XXXXXX), you can use lookbehind to target the number right after that timestamp:

(?<=^\d{2,}/\d{2}/\d{2}\s\d{2}:\d{2}:\d{2}\.\d+\s)(\d+)

Breakdown:

• (?<=...): Positive lookbehind, checks that the text before our target matches the timestamp pattern
• ^\d{2,}/\d{2}/\d{2}\s\d{2}:\d{2}:\d{2}\.\d+: Explicitly matches the date and time part
• \s: Matches the space after the timestamp
• (\d+): Captures the target number

Option 3: Exclude smaller numbers (if your target is always the longest digit sequence) #

In your example, 1245680 is the longest sequence of digits. If that's always true for your input data, you can directly target 7-digit numbers:

\d{7}

Alternatively, if you're working in a programming language, you could collect all \d+ matches and pick the longest one—though this relies on the target always being the longest digit sequence.

Whichever option you choose, be sure to test it against your full set of input strings to confirm it works consistently!

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