How do I search for lines containing a string that are followed by a line that does NOT contain that string?

Question

I have a text file with lines like the following:

3005,23,1,NI,2348,2
3005,17,NI,2348
3005,23,1,NI,2348,2
3005,17,NI,2347

The 2348 and 2347 in those lines could actually be any numbers.

I want to find the lines starting with 3005,23 that are followed by a line starting with 3005,17 where the 5th value on the first line matches the 4th value of the second line (for example, 2348 in the first 2 lines of the example). I know how to do that using a capture group:

/3005,23,[0-9],[A-Z][A-Z],\([0-9]\+\),.*\n3005,17,[A-Z][A-Z],\1

That search will match the first 2 lines in the example above.

But how do I search for the lines starting with 3005,23 that are followed by a line starting with 3005,17 where the 5th value on the first line does not match the 4th value on the second line, so that it matches the last 2 lines in the example above?

Answer 1

I would do:

/3005,23,[0-9],[A-Z][A-Z],\([0-9]\+\),.*\n3005,17,[A-Z][A-Z],\1\@![0-9]\+

\@! match with zero width only if the preceding atom do not match.

More information width :help /\@!