MobileRead Forums - View Single Post

kiwidude · 05-14-2011, 02:14 PM

That's just the way it works now.

Calibre merges together books that are all in the same "ISBN pool" from across your sources picking what it considers the "best" choice for title, author, series etc from all of those records. So if you search Amazon and Goodreads then if their ISBNs are the same or recognised as belonging together according to the XISBN website then you will only see one combined result displayed.

You will see multiple books appearing if the search results return books that have ISBNs that are not in the same pool. Say for instance you searched for a book with "Dragon" as the title (without an ISBN already known), and the author had multiple books with "Dragon" in their title. Then they will come back as your multiple results to choose which "book" you were after.

So that is very different to presenting a result per source like I think you got prior to 0.8 (I never used it). The only real downside is that it means you cannot pick and choose on the fly from book descriptions from different sources or some other attribute you prefer from one (not without doing the search over again and disabling populating certain fields from certain plugins). Instead you just have to trust that the merge logic is grabbing a result that you will like.

It woud be a very complex slippery slope if Kovid was to try and support that though, for instance what if Amazon had a description but no series, but Goodreads had a series name. Letting Calibre merge the two gives you it's opinion of the best of both. You can always manually overwrite it if you want to using the book id hyperlinks to quickly navigate to the various site specific editions.

05-14-2011, 02:14 PM	#2
kiwidude Calibre Plugins Developer Posts: 4,732 Karma: 2197770 Join Date: Oct 2010 Location: Australia Device: Kindle Oasis	That's just the way it works now. Calibre merges together books that are all in the same "ISBN pool" from across your sources picking what it considers the "best" choice for title, author, series etc from all of those records. So if you search Amazon and Goodreads then if their ISBNs are the same or recognised as belonging together according to the XISBN website then you will only see one combined result displayed. You will see multiple books appearing if the search results return books that have ISBNs that are not in the same pool. Say for instance you searched for a book with "Dragon" as the title (without an ISBN already known), and the author had multiple books with "Dragon" in their title. Then they will come back as your multiple results to choose which "book" you were after. So that is very different to presenting a result per source like I think you got prior to 0.8 (I never used it). The only real downside is that it means you cannot pick and choose on the fly from book descriptions from different sources or some other attribute you prefer from one (not without doing the search over again and disabling populating certain fields from certain plugins). Instead you just have to trust that the merge logic is grabbing a result that you will like. It woud be a very complex slippery slope if Kovid was to try and support that though, for instance what if Amazon had a description but no series, but Goodreads had a series name. Letting Calibre merge the two gives you it's opinion of the best of both. You can always manually overwrite it if you want to using the book id hyperlinks to quickly navigate to the various site specific editions.