Parsing pageranges

Hi, I have several systems running which gather, store and process bibliographic data. I have treated pageranges on the basis of two fields per range - start and end, with supplemantary ranges available as well. I've never had to deal with more than 3 ranges in a reference: chapter, references and notes. Now, a client has asked for 6 ranges. I could futureproof this by putting 10 ranges in but I got to thinking about how Pagemaker used to handle prining (and how Micro$oft do now) where I can note a range as (example) 1-4,6,8-10. I need to be able to parse this type of string so that I can identify the number of pages being referenced. I also need to ensure that the user hasn't entered a mixed range such as xiii-5 (I know that the second part of that is 1-5 but I don't know what the highest roman numeral was). I do know how to handle the roman calculations, so that's a side issue. I'm guessing that regex is the way to go, but whenever I'm confronted with it, I look for a chinese interpreter. Any suggestions? George in Oxford