Thanks, that fix helped me instantly fix the parsing of a 55k lines log file. I really appreciate your prompt help. Your fixed version makes a lot of sense. I had been dabbling with putting the "random" in the record and also in the main "records" sequence, but I never came across the `+(record | random)` variant in my experiments :)
This is just beautiful.. thousands of lines parsed in a snap! Thanks @zevv for this wonderful package. (Next step: parse the records to JSON and then to HTML.)
