View Single Post
Old 10-11-2015, 09:07 PM   #12
fidvo
Addict
fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.fidvo ought to be getting tired of karma fortunes by now.
 
Posts: 309
Karma: 1645952
Join Date: Jun 2012
Device: none
Quote:
Originally Posted by OldLincoln View Post
I loved the idea but it doesn't consistently solve the problem. For example:
"Danny Logan Mystery [2]: Danny Logan 2: No Way to Die"

becomes:
"Danny Logan Mystery [2]: Danny Logan 2:"

It takes consistent inputs to get consistent results. We'd need a strong AI to work through the garbage naming lack of convention on Amazon, not to mention the changing of minds with changing names in series, titles, identifiers and authors. Fixing that isn't going to happen so I'll keep what i got till something works better.

Thanks for your help.
The second rule of Calibre is this:

Downloaded metadata is just a suggestion. In your example, Amazon helpfully broke the first rule by adding the Series to the Title field. That's why you should always review the metadata after downloading, and clean it up when necessary.

Someone with more experience with me might be able to help you construct a custom search for all books containing a number somewhere in the title. That could help narrow down your library to books likely to need the title field to be cleaned. If there aren't too many, you could go through them by hand and fix them. Otherwise, the proper regex in Search-and-Replace might work.
fidvo is offline   Reply With Quote