Home / Questions / 384320
Accepted
Pablo Lozano
Asked: 2020-08-25 00:42:18 +0800 CST 2020-08-25 00:42:18 +0800 CST

Capture groups in regular expression conditioned by the following characters

  • 772

I am grouping records (tasks) that have an identifier code.

The code is unique, but there are tasks that are a continuation of others and this is reflected in its code. For example:

  • 12345: Original record
  • 12345A: New version of register 12345
  • 12345B: Another version of register 12345
  • 12345C: Third version of register 12345

And here comes my problem:

  • 12345CONT: Continuation (of the active version) of 12345 (it does not matter if the active one is the original or A, B or C)
  • 12345CONT2: Second continuation of 12345
  • 12345PF: Final Information
  • ... (other endings, always more than one letter

I want to capture the following groups:

ID - Versión (A,B,C,D -opcional-) - Tipo de tarea (CONT<n>, PF<n>, ...)

My regex is currently the following:

(\d+)[ABCD]?(CONT\d?|PF\d?|REV|REP)?

But it has the problem that 12345CONTit captures 12345 - Cme and discards the ONT. I'm looking at "look-ahead" options, but I haven't quite found the solution.

How to capture CONT and C separately?

java

1 Answers

  1. Best Answer

    You can use 'negative lookahead' :

    (\d+)(A|B|C(?!ONT)|D)?(CONT\d?|PF\d?|REV|REP)?

    C(?!ONT)will capture Conly when not followed by ONT.

    ( demos )

    • 7