正则表达式-字符串中任何地方都不包含某个字符串

Question

I hope this is not a duplicate, I have been searching reg ex answers for hours so I don't think so!

Is there a way of searching with regular expression to say the result must not contain the word 'bot' anywhere in the string?

I am using a negative lookahead elsewhere but the problem is I don't know where the word 'bot' may appear. It could be part of a longer word, it could appear at the start, at the end, or both! I want to include results that match the rest of my expression but also exclude 'bot'.

This is my current search expression:

2017-0[2-5] (.*) (UserInfo.aspx UID=111 80 -)=(?!10. *)+. * 

And the results I am searching on generally look like this:

2017-02-16 15:56:00 10.3.1.17 GET /UserInfo.aspx UID=111 80 - 157.85.39.17 Mozilla/5.0+(compatible;+bingbot/2.0;++http://www.bing.com/bingbot.htm) 200 0 0 390

Ideally I would want this result to be excluded as it contains 'bot' even though it still matches the first part of my RegEx.

Answer 1

Implement this as a two step process if possible. If, for example, this is in a script, have two checks: one that tests whether a line contains bot, and then a second step that performs your existing regex, if bot was not there.

Breaking a complex matching requirement into multiple tests is often a good idea. It results in clearer code, and it also may be more efficient, especially if the earlier checks are simple.

Breaking this step out separately also allows you to use a simpler positive match instead of a negative match.

Example pseudo-code:

if (!string.contains('/bot/') and string.contains('/yourpatternhere/'))
{
    do stuff
}

If you are restricted to a single regex, simply use a negative lookahead at the beginning of the string:

(?!^.*bot)2017-0[2-5] (.*) (UserInfo.aspx UID=111 80 -)=(?!10. *)+. * 

(I'm ignoring any other possible issues with the regex here, which doesn't seem quite right...)