til/unix/use-negative-lookbehind-matching-with-ripgrep.md

Use Negative Lookbehind Matching With ripgrep

The most straightforward way to use ripgrep is to hand it a pattern. It will take that pattern and move forward through each file trying to find matches.

$ rg 'TwilioClient\.new'

That will find all occurrences of TwilioClient.new in available project files.

What if that pattern is too permissive though? That is going to match on occurrences of TwilioClient.new as well as things like LoggingTwilioClient.new. If we want to exclude the latter, there are a few ways to do that. One of them being the use of the negative lookbehind regex feature that is available with PCRE2 (Perl-Compatible Regular Expressions).

A negative lookbehind is like a standard pattern. We look forward through the document for the base pattern (like TwilioClient\.new). However, once we find that match, we then look back at the previous characters and if they match our negative lookbehind pattern, then it is no longer a positive match.

We can use one of the following to forms to achieve this:

  (?<!...)                    )
  (*nlb:...)                  ) negative lookbehind
  (*negative_lookbehind:...)  )

For instance, here is what this looks like for our example:

$ rg -P '(?<!Logging)TwilioClient\.new'

Note: we have to use the -P flag to tell ripgrep that we are using PCRE2 syntax. Otherwise, it assumes a simpler regex syntax that doesn't support negative lookbehind.

See man rg for more details.