Add Books, one record per format?

[unboggling]

Anyone know of a way to Add Books in bulk so that the result is one format per record for all Title dupes and for all Title/Author dupes? Automerge options turned on don't seem to do it. Automerge off doesn't do it. Is there anything in calibre command line that will let me do it?

In other words I want each incoming format to have a separate book record regardless of Title dupes or Author/Title dupes.

(Search of the calibre manual didn't produce anything useful.)

[unboggling]

Maybe there's a different or better way other than one format per record. What I want to do is import a bunch of books then evaluate each format using the calibre viewer or other reader with OpenWith plugin, then keep the best format and delete the others. It just seems to me to be faster and easier if there is only one format per record but maybe I'm missing something or using bad assumptions?

[kiwidude]

@unboogling - I find the faster way to do this is to definitely not bring them all into calibre, instead only the format version you want to keep.

If you associate the calibre ebook viewer with file extensions like epub (as well as whatever other readers you prefer configured like Microsoft Reader for LIT files etc), you can do all the comparisons in windows explorer just by double clicking on the file name. Much faster imho than importing and then deleting in calibre.

I use windows explorer search to lookup an author and/or title name in my "unsorted" folders, then just directly double-click on the search results. Once I have my preferred format to import only then do I stick it in calibre. That way you have other information visible to you like file size etc to help dictate your ordering of "likelihood" of a best version, duplicates etc. You also don't have to look at every format either.

[unboggling]

Originally Posted by kiwidude

@unboogling - I find the faster way to do this is to definitely not bring them all into calibre, instead only the format version you want to keep.

Thanks. What I'd been doing is importing a few at a time, then evaluating them in calibre, then only importing and evaluating other formats if the first wasn't good.

Your way seems most efficient but also requires going only a few at a time. I should probably start evaluating formats in the OS first before import to calibre, a few at a time. But for a large bulk import in which there are usually multiple formats of a title, where the authors are sorted FN LN by the OS, I was thinking I could import in bulk, then without making many changes except standardizing Author/Series/Title, Save Out so that the books would be sorted by LN FN. I haven't figured out a way to use a file renamer across all the Author directories at once in bulk to do the switch to LN, FN in the OS. (Also, one of the drawbacks to Adding to calibre then Saving out is losing the original OS file-modified date in the Save.)

Originally Posted by kiwidude

I use windows explorer search to lookup an author and/or title name in my "unsorted" folders, then just directly double-click on the search results. Once I have my preferred format to import only then do I stick it in calibre. That way you have other information visible to you like file size etc to help dictate your ordering of "likelihood" of a best version, duplicates etc. You also don't have to look at every format either.

In the case of this particular source folder, I don't like working with the OS's FN LN sort when calibre's sort is LN FN. It's time consuming to orient and switch back and forth between the the different sorts. Searching the Source folder by author in OS is what I normally do for most Source folders. But in this one there are multiple duplicates of variously named authors and titles across differently named author folders and title folders throughout this Source folder. So evaluating in the OS doesn't seem as efficient as doing it in calibre where some time consuming bulk metadata edits will at least eventually put all the copies of the same book into conjunction in the calibre sort or search, while evaluating in the OS would not have them in conjunction. EDIT: The Source folder is a mess regarding naming conventions. An OS search doesn't find everything. Some authors names are missing. Some use series instead of author. File names for title dupes are not necessarily the same, some include author, some include series, some include abbreviations for title, some don't include entire title.

I'll probably have to go back to the few-at-a-time method. Evaluate out in the OS as you do. But I wanted to find a way to group all author/title dupes from across the Source folder first, which seems to require a preliminary evaluation and metadata edit of Author, Series, and Title.

[kiwidude]

I think you have contradicted yourself in your answer. First you say that you have an author name good enough that you can import it into calibre. Then you say that you cant use windows search because the author name doesn't always exist. Now unless you plan on relying on some metadata in the file when you bulk add to calibre (which you can never rely on imho, and it is only present in some formats etc) then you have the same "problem" either way?

If your input directories contain completely meaningless filenames, then your options aren't great regardless.

I make sure that as I get new books they always have a name I can later search on. Preferably in every filename, but worst case on the parent folder. Windows search can find either situation (unlike adding to calibre), it just means you have to drill into the author name folders to scrape up any books that didn't match the search.

Yep it is not a quick process. But unless you plan on buying every book all over again and putting in calibre immediately then I don't see any alternative. Had I been disciplined in rationalising my books as I accumulated them over the years as some people are, I wouldn't have an issue. Instead it was a "sort it out one day" job. Well one day has arrived. Hence why as I posted in your other thread ages ago I am just taking an author by author approach, and most days come home to chip away at a few more. I'm in no major hurry - I've got enough processed books to read for at least the next 10 years already

There is no magic bullet to make bulk importing happen by magic whereby calibre gives you this perfect metadata duplicate free best possible format version of a pile of kak input. We would all love there to be such a thing, but it doesn't exist. Garbage in, garbage out.

[unboggling]

Sorry if I wasn't clear and sounded contradictory. Thank you for your responses.