Begin Edit: Sorry, Kovid, I read your post after answering. However, I believe this can stand as a final note on regexes in this thread
End Edit
Without answering every one post about regexes individually, I'd like to at least make two points about those darn things:
First, you need to understand that computers (generally) lack the semantic understanding of text that we humans have. A computers sees an ebook not as a collection of words gathered into sentences that convey meaning, but as a string of characters- a string. Thus, you need to understand that regular expressions also only operate on strings- there is no way you can get the computer to understand what you're trying to do on the level of words or meanings, you have to tell it exactly what to do on the level of essentially manipulating single characters. This is, in particular, aimed at Elfwreck (and not intended as a criticism, only as something I got the feeling you didn't quite realize).
That said, my second point is this: To all those here complaining about regular expressions, I'd like to ask you to read
the tutorial that's available in the manual. (Better yet, read
the first post of the original thread, since that is likely to become a sticky if the changes proposed here are implemented- although I haven't gotten around to update that to include the changes made in 0.7.42.) The goal of that tutorial is to take the user from the point of knowing nothing about regexes, but being willing to learn, to the point where he (or she) is comfortable enough to use them in Calibre. Bearing that in mind, please tell me what should be improved about it- preferrably relatively precise, as I don't know how to process the information that "it is just too difficult" or something. If I know what to improve, I'll try to do that, although it may take some time. (To avoid derailing this thread, I'd suggest you PM me.)