First, sorry for jumping in here - Do ignore me and this babble if I have read and understood this the wrong way!!
But.. To my knowledge this regex-excersice sounds reversed. Whatever divides the musicbrainz ids (whatever that is..) should
not matter as what needs to be identified is one or more blocks matching the pattern dcb03ce3-67a5-4eb3-b2d1-2a12d93a38f3. (Something like (\w{8}-\w{4}-\w{4}-\w{4}-\w{12}) - Why are you discussing the dividers when they don't matter?
Also, and a slightly unrelated rant, sorry again - skip if you like - what the f.. is this musicbrainz new-kid-on-the-block latest not-human-readable no-thankyou stuff? Well, to be honest, I don't really care as long as you aren't forcing it down my throat!? I have thousands of albums and a re-scan will take forever! Also, Is it part of any organization or just some random dudes hacking something up? Im sorry but storing data in my files that is simply a (temporary) key to some obscure site of the internet goes against all common sence. The storing of hard meta-data is cool, but storing 235235-235-6-4-24-6-4-26-42-64-6246-whoop-de-doo.. Really..!?! Have you all lost it??
What I'd like to know is where is the missing artwork-designer, producer/mixer etc.. tags, and where is the good meta-data? proper release dates (yyyy-mm-dd) and proper genres please. ("oldies" and "pop" is not genres). Why not just get it all from wikipedia? And if the autobots are too stupid to find the proper page (really easy task for 99% of the time), ask the user if multiple options are available.
Peace all - and thanks for your hard work