[GUI Plugin] Find Duplicates

[kovidgoyal]

IIRC calibre does read language metadata from both epub and mobi formats.

[DoctorOhh]

Originally Posted by kiwidude

If indeed this is happening, then I think the most appropriate solution to all this is to add a preferences option for this plugin as I originally intended, with the default being off. So if your languages field gets filled "supriously" you don't care, and if it gets populated religously you have an option to make use of it. In the meantime you can always use the bulk metadata editor to clear the field.

However that is all speculation - it is the only thing that changed in the plugin and until/unless someone verified the above is happening to them and having languages set on their books unintended then I'm not going to change anything. Personally I *never* turn on import metadata from the book (only filename), and equally I *never* try to download the languages field via metadata download. However perhaps others are getting caught out by this.

A preference option with the default being off might be best. I just looked at my library of 9866 books and I now have 174 English and 5 French. All of the books are English and I never set the language on any of them.

I just used the edit metadata window and removed all languages so I'm good, but I never noticed this because I hide that item in the tag browser.

[kiwidude]

Changes in this release:

• Make the languages comparison optional (default false) via a checkbox on the Find Duplicates dialog

[JayKindle]

I've installed this plugin to 0.8.29, but don't see it in the GUI. How can I run this plugin? I've restarted calibre, but it doesn't seem to appear anywhere, though in the plugins section, it states that it is enabled or at least installed.

[kiwidude]

@JayKindle - sounds like you skipped the step when installing the plugin of indicating which toolbar(s) you want the plugin to appear on. Go to Preferences -> Toolbar and configure it there.

[Noughty]

Is it possible to include the series info into the search? Sometimes there is a series with the same title, so it comes up as a dupe. Also a problem with serials which have the same title for every chapter.
For example if the series field has different series index it shouldn't count it as dupe. Maybe something like a check box - a setting you can choose if you want. Since there might also be a dupe and incorrect series field (so it might not be found).
Just a suggestion. I wish authors just add the number to the title, now there might be 12 different works with the same author and title.

[kiwidude]

@Noughty - I guess I've never come across this situation, I certainly in my library don't have any books where the title is the same across the series. I guess if someone was storing magazines and using the tiele as the magazine name and the series index for the year/month that could happen.

However generally speaking reusing the same title for the same author (as in the situation described) is not a good thing to be doing imo. One reason that comes to mind is that every time you add another in the series, calibre is going to prompt you with its duplicate dialog, since it matches based on title and author. Likewise automerge uses only title and author as well.

For those reasons I think you are better off steering away from it if possible. From a Find Duplicates perspective, it is easy enough to just add that set of books to a duplicate exemption group, then you won't ever see them appear again anyways?

[Noughty]

Yes, I can set an exemption, but it's still troublesome. As you pointed out with adding the books. They aren't magazines, just books. And a few books released as serials, like a chapter every week. They find it too complicated to add chapter number to the title. Will have to fix it manually, since I already lost some of the chapters even while adding, since it's the "same" book. That's why I never delete the originals . Forces me to have 3 files instead of one. One original, one calibre and one calibre back up.
So as you pointed out the biggest problem is while adding them, which isn't about this plug in.

Your goodreads metadata plug in helps, since these chapters have their own goodreads entry and I can link it correctly.

[theducks]

Noughty
You might consider building an (ongoing) Omnibus with Sigil (or other editor).
Every issue period, you add that to The Omnibus kept in Calibre.

Open With Plugin makes this easy to launch. Sigil has an 'Add Existing' (file) feature.

[grantshoarma]

Hello,

After finding binary duplicates, is there a way to delete all redundant files in bulk, rather than selecting each one by hand?

[theducks]

Originally Posted by grantshoarma

Hello,

After finding binary duplicates, is there a way to delete all redundant files in bulk, rather than selecting each one by hand?

I think you need to decide Which of the duplicates NOT TO KEEP/or to merge. Both are shown.

[grantshoarma]

I think you need to decide Which of the duplicates NOT TO KEEP/or to merge. Both are shown.

I understand the reason to do this for e.g. duplicate titles, but for a file that is an EXACT replica of another, I don't see the harm in giving me a "merge all" option. Or perhaps I am misunderstanding the definition of a binary duplicate.

[Noughty]

I was thinking of an automatic selection (when all dupes are shown at once) of dupes, leaving only one file (choosing could be on the filled metadata but not necessary). You can go through the list and unclick/click if you don't like the automatic choice. I don't know about these things, but maybe it's not hard to make it select one of the dupes?
Most often with dupes I just delete one of them since they are usually duplicate file, very rarely I need to merge or do some other action. So it would be a nice option - to select, not to delete, since adjustments might be needed on which entry to delete.

[kiwidude]

What you can do "in bulk" depends on whether you want to merge or delete.

If you are wanting to merge the metadata between the book rows, then no you will have to individually select the pairs and merge them together.

If however you just want to remove the binary duplicate formats, then you should select the "Show all groups at once with highlighting" option. Then control-click on one book from each group representing the book to remove, and then right-click to remove them.

[grantshoarma]

I just want to delete the duplicates, but for some ridiculous reason I have thousands of them, translating to thousands of Ctrl-clicks........ As noughty says, if there was an option to "select all duplicates," it would be useful.

So if I had 3 copies of a book, 2 books from that group would be selected. Then I could right click and delete.