You can create a microflow that looks something likes, you can click on it for a larger version. When you have any more questions about the flow leave a comment and i'll give a better explanation. But this will hopefully explain enough.
If the length of 'group1' is always the same you can use the substring function.
So if the length of 'group1' is always 5 characters you can use substring(group1), 5
But as csv-files are ussualy not built up this way (so with varying 'group length') I think the easiest is to resort to Java and let Java return the separated string values