Regex: match 8 character words containing at least 3 instances of a digit or a specific symbol

I'm trying to create a regex that matches 8 character words that contain at 3 least instances of a digit or the '%' symbol and whitespace after the word.

The following strings should match:

• 'ab1d2f3h '
• 'ab%d2f3h '
• 'ab%d%f3h '
• 'ab%d%f%h '

The regex I have so far looks like this:

const string = 'this word ab1d2f3h needs to go, but not this word abcdefgh %%%'.replace(/(?=(?:\D*(\d|%)){3})(\w|%){8}\s/g, '%%%%%%%% ')

If I remove '%%%' from the string, it works - 'ab1d2f3h ' is replaced. However, if '%%%' is present in the string, it also replaces 'abcdefhg ', which I don't want to happen.

Does anyone know the proper regex for this?

If a lookbehind assertion is supported in the environment where you are using the pattern:

 (?<!\S)(?=\S{8}\s)(?:[^\s\d%]*[%\d]){3}\S*

Explanation

• (?<!\S) Positive lookbehind, assert a whitspace boundary to the left
• (?=\S{8}\s) positive lookahead, assert 8 non whitespace chars followed by a whitespace char (note that it can also match a newline)
• (?:[^\s\d%]*[%\d]){3} Repeat 3 times matching any non whitespace chars except a digit or % and then match either a digit or %
• \S* Match optional non whitespace chars

Regex demo

Or using a capture group:

(^|\s)(?=\S{8}\s)(?:[^\s\d%]*[%\d]){3}\S*

Regex demo

In the replacement use the first capture group to keep the leading whitspace char.

<!-- begin snippet: js hide: false console: true babel: false -->

<!-- language: lang-js -->

const string = 'this word ab1d2f3h needs to go, but not this word abcdefgh %%%'.replace(/(^|\s)(?=\S{8}\s)(?:[^\s\d%]*[%\d]){3}\S*/g, '$1%%%%%%%%')
console.log(string)

<!-- end snippet -->

Asked ChatGPT and it answered my question:

You can use the following regular expression to match 8 character words containing at least 3 instances of a digit or the symbol '%':

\b(?=(\w*[\d%]){3})\w{8}\b

Explanation:

• \b matches a word boundary to ensure that the regex only matches whole words.
• (?=(\w*[\d@]){3}) is a positive lookahead assertion that checks for the presence of at least 3 instances of a digit or the symbol '@' in the word.
• \w{8} matches any 8 word characters (letters, digits, and underscores).
• \b matches the end of the word boundary.

Note that this regex assumes that there are no spaces or other non-word characters in the string. If there are, you may need to modify the regex to handle them.