MobileRead Forums - View Single Post

kiwidude · 06-01-2012, 08:21 AM

Make sure that you only have metadata download plugins enabled that you *really* want data from, don't just run with all of them or even necessarily all the default ones. Just pick a couple of books, do the download, look at the log with the time taken and decide whether you really want the particular output from that plugin or not. If you use the Amazon plugin, make sure it is set for your most local location. The Overdrive plugin is extremely slow so don't use that if you are in a hurry. There are also other alternatives from the plugins subforum such as Goodreads, Fantastic Fiction & B&N and others.

Many of the plugins will run faster if you have the ISBN for the book populated already. Normally I would point you to the Extract ISBN plugin for that, but if you are using calibre 0.8.53 or 0.8.54 at the moment that plugin is broken, and possibly won't get fixed until calibre 0.8.55 comes out.

Personally from experience I don't use bulk metadata download *especially* if you do not have the ISBN for a book. By doing so you are rolling the dice that the results returned from a website are ordered to exactly match the right book. Too often they are not, it is not the plugins fault when this happens, it is the website's own algorithm for finding a best match for that title, which can go *really* wonky if they don't actually have that book under the same name. And if you have overwrite title and author turned on in your metadata options, you are going to end up with an utter mess in your library of book entries having different books to the title/author in the library.

So I do individual metadata downloads to choose exactly which results I want, and frequently have to run it multiple times to get the matches across all the websites I want linked to. Does it take longer? Sure, but I'm not exactly going to be reading 9000 books in the next few weeks either. And at least I am 100% confident it will not make a mess of my library by pulling metadata for the wrong books and screwing up my library, which means much more to me in the longer term.

06-01-2012, 08:21 AM	#4
kiwidude Calibre Plugins Developer Posts: 4,637 Karma: 2162064 Join Date: Oct 2010 Location: Australia Device: Kindle Oasis	Make sure that you only have metadata download plugins enabled that you really want data from, don't just run with all of them or even necessarily all the default ones. Just pick a couple of books, do the download, look at the log with the time taken and decide whether you really want the particular output from that plugin or not. If you use the Amazon plugin, make sure it is set for your most local location. The Overdrive plugin is extremely slow so don't use that if you are in a hurry. There are also other alternatives from the plugins subforum such as Goodreads, Fantastic Fiction & B&N and others. Many of the plugins will run faster if you have the ISBN for the book populated already. Normally I would point you to the Extract ISBN plugin for that, but if you are using calibre 0.8.53 or 0.8.54 at the moment that plugin is broken, and possibly won't get fixed until calibre 0.8.55 comes out. Personally from experience I don't use bulk metadata download especially if you do not have the ISBN for a book. By doing so you are rolling the dice that the results returned from a website are ordered to exactly match the right book. Too often they are not, it is not the plugins fault when this happens, it is the website's own algorithm for finding a best match for that title, which can go really wonky if they don't actually have that book under the same name. And if you have overwrite title and author turned on in your metadata options, you are going to end up with an utter mess in your library of book entries having different books to the title/author in the library. So I do individual metadata downloads to choose exactly which results I want, and frequently have to run it multiple times to get the matches across all the websites I want linked to. Does it take longer? Sure, but I'm not exactly going to be reading 9000 books in the next few weeks either. And at least I am 100% confident it will not make a mess of my library by pulling metadata for the wrong books and screwing up my library, which means much more to me in the longer term.