A single field (description) in the database is being returned as a large string
You may or may not be able to pull this off by nesting a bunch of the "TEXT" functions built in to jaspersoft studio, combining 'substitute',and 'search' to whittle this string down to everything between the third and fourth instances of '-', but I don't recommend it. If it is possible, it's not easily maintainable. What you could do is add apache's commons-lang jar into your project, split that string (using, no surprise, the StringUtils.split() method) on the '-' hyphen char, and return the appropriate string from the resulting array list. The split method can split your string into an array list of strings based on a given character.
Also, to be fair, Java also has a string split method, but it's not as intuitive:
And finally, I'm sure you can do this via regex as well, but again it's not as intuitive as the apache commons-lang approach: