Quote:
Originally Posted by lumpynose
So is it correct to say that the 3 likely scenarios are, 1) the book has no cover in the epub and calibre generates one from the initial text, 2) the book has a cover image, but it's not in an xhtml file that's tagged as the cover / title page, and 3) the book has a cover image and is in the cover /title page. With scenarios 2 and 3 calibre extracts the cover image and displays it when you're using its cover grid mode.
|
That's basically right. When you add the book, or press the set cover button in the metadata editor, calibre examines the book and tries to get a cover. There are a few methods which include "image with name cover". Or "image in first file". I don't know if there is a difference in the methods used when adding a book or pressing the set cover button.
Quote:
Scenarios 1 and 2 are where the margins are messed up but the epub can be transformed into scenario 3 by running Polish books or Modify ePub.
Short of opening each epub in calibre's editor, is there an easy way to differentiate between an epub in scenarios 2 and 3? Or can I run Polish books on all of them and ignore whatever state they're in? (Probably not a bad idea in general, in order to smarten the punctuation and other stuff.)
|
Running Polish books or Modify ePub do not harm anything. My process is usually to update the metadata and cover when I add the book to my library. And then send it to the device. If the cover doesn't show up, I'll fix it. Otherwise, I'll probably leave it until I read the book. Then I fix whatever is needed. It does depend on other things. If I find a particular series/author/publisher/source have problems, I'll fix them before putting them on the device. For example, books from Smashwords are usually badly constructed. I usually fix these soon after adding them to my library.
In any case, even when I do fix things when adding, I am not adding enough books for it to be a pain.