Split Library up and have some regrets

[RandomLibrarian]

I recently split up my library of approx. 4000 books in to different libraries because getting a good solid grasp on fixing the metadata was overwhelming. It worked wonders and I've got about 80% of the cataloging done. However, there have been a few times recently I ended up searching different libraries to make absolutely sure I didn't have a book I could have sworn I did. No biggie, but this led me to dive into options about searching multiple libraries at once.

I discovered that virtual libraries wouldn't help me here. I learned that Calibre actually recommends you don't do what I did, and split your libraries. This is where frustration kind of reared its head. I mostly agree with Calibre, and when I am done cataloging all my items maybe I will bring them all back together. The problem I have is that the limitation seems to exist in a vacuum that doesn't fit reality. The find metadata feature of calibre kind of sucks if you are going for a relatively accurate set of metadata. It'll get close, but it will mess up almost as much as it fixes and you really won't save much time over doing it manually for small to medium sized libraries.

Would be nice to see a feature to bring physically separate libraries together via virtual libraries or linked libraries. (This isn't a feature request post) Has anyone overcome this kind of hurdle and kept their libraries separate? I'd like to hear feedback by people who've worked through this process before so I don't attempt reinventing any wheels. Thanks.

[theducks]

IN CALIBRE
Select books in the Library you no longer want: right-click:Copy to Library (choose option:delete is typical):<destination Library>

[Sirtel]

Why would you split up 4000 books? That's a small library. I have over 20 000 all in one library. True, I've fixed metadata and formatting as I buy them, but there have been periods I haven't had the time and accumulated a queue of hundreds of books. Still, I fail to see how splitting the library would help anything. I just sort by date, newest first, and start on top.

[DNSB]

My workflow uses two libraries. One library (mine is called Intake) gets newly added ebooks which need metadata, covers, editing plus whatever else I want to do to modify them, the second library with (mine is called Main) gets the ebooks after they have been massaged. Main is where I send to my ereaders, etc.

And yes, the library names are rather obvious and boring.

When I was first moving to using calibre, I used some bulk metadata downloads but there were always problems with a goodly percentage of the ebooks so I now use individual metadata downloads. This takes a bit longer than the bulk download but less time overall since I don't have to redo the bad metadata downloads.

[ownedbycats]

I used to have separate libraries for my eBooks, physical books, fanfiction, and documentation/manuals. Now I just use virtual libraries. I use View Manager plugin extensively to swap between different column sets.

I think the CalibreSpy plugin lets you view multiple libraries at the same time, though its read-only. There's also another plugin by the same author that helps in merging libraries together but the name of it's escaping me right now.

[Bookstooge]

Ahhh, ma cherie, don't we all have SOME regrets?

[hobnail]

I agree that being able to virtually merge/combine multiple separate libraries would be nice. I have two separate libraries. One contains books on my ereader or destined for it (books that need their css and html massaged have a No in the cssFixed column and aren't yet on the ereader). I can connect the ereader and sort by cssFixed and see which ones need to go on it, first taking care to do the copy+delete thing that theducks pointed out and copying to the other library the ones that I've finished reading and deleted from the ereader. So the other library is for books I've read.

I tried using virtual libraries, one for unread and the other for everything, finished and unread, but whenever I started calibre it would start with the everything one which annoyed me. (I think that's what was happening, or whatever it was it was vexing.)

[ownedbycats]

Originally Posted by hobnail

I tried using virtual libraries, one for unread and the other for everything, finished and unread, but whenever I started calibre it would start with the everything one which annoyed me. (I think that's what was happening, or whatever it was it was vexing.)

Preferences > Behaviour > "Virtual library to apply when the current library is opened"

View Manager also has a similar option in its preferences (for views, which can be attached to specific VLs) and it'll override the Calibre settings if it's set.

[hobnail]

Originally Posted by ownedbycats

Preferences > Behaviour > "Virtual library to apply when the current library is opened"

View Manager also has a similar option in its preferences (for views, which can be attached to specific VLs) and it'll override the Calibre settings if it's set.

Thanks. I forgot or, more likely, overlooked the View Manager. It looks like what I need when using virtual libraries. I'll try combining my two libraries and see how it goes.

[Sirtel]

Originally Posted by DNSB

My workflow uses two libraries. One library (mine is called Intake) gets newly added ebooks which need metadata, covers, editing plus whatever else I want to do to modify them, the second library with (mine is called Main) gets the ebooks after they have been massaged. Main is where I send to my ereaders, etc.

And yes, the library names are rather obvious and boring.

When I was first moving to using calibre, I used some bulk metadata downloads but there were always problems with a goodly percentage of the ebooks so I now use individual metadata downloads. This takes a bit longer than the bulk download but less time overall since I don't have to redo the bad metadata downloads.

My problem with that approach is that while editing metadata I want to be able to see other books by the author I already have (to set the correct series info, for example). Can't do that with separate libraries.

[ownedbycats]

Originally Posted by Sirtel

My problem with that approach is that while editing metadata I want to be able to see other books by the author I already have (to set the correct series info, for example). Can't do that with separate libraries.

My personal solution to that is instead of an intake library, I use an intake virtual library. It's set to filter the @cleanup tag and my other VLs hide that tag. Then I can use quickview to reference other books.

As mentioned earlier CalibreSpy also lets you read-only access of other libraries.

[theducks]

If you have 2 Monitors (or 2 PC's. but no Copy & paste this way)
I use the Content Server to review other books (and reference Web Pages)

[JSWolf]

Originally Posted by RandomLibrarian

I recently split up my library of approx. 4000 books in to different libraries because getting a good solid grasp on fixing the metadata was overwhelming. It worked wonders and I've got about 80% of the cataloging done. However, there have been a few times recently I ended up searching different libraries to make absolutely sure I didn't have a book I could have sworn I did. No biggie, but this led me to dive into options about searching multiple libraries at once.

I discovered that virtual libraries wouldn't help me here. I learned that Calibre actually recommends you don't do what I did, and split your libraries. This is where frustration kind of reared its head. I mostly agree with Calibre, and when I am done cataloging all my items maybe I will bring them all back together. The problem I have is that the limitation seems to exist in a vacuum that doesn't fit reality. The find metadata feature of calibre kind of sucks if you are going for a relatively accurate set of metadata. It'll get close, but it will mess up almost as much as it fixes and you really won't save much time over doing it manually for small to medium sized libraries.

Would be nice to see a feature to bring physically separate libraries together via virtual libraries or linked libraries. (This isn't a feature request post) Has anyone overcome this kind of hurdle and kept their libraries separate? I'd like to hear feedback by people who've worked through this process before so I don't attempt reinventing any wheels. Thanks.

The thing to do is create a new library using the template of the current library. Then move the eBooks (that need fixing) from the current library to the new library. Fix the eBooks in the new library and move them back to the current library. That will work no problem.

[JSWolf]

Originally Posted by DNSB

My workflow uses two libraries. One library (mine is called Intake) gets newly added ebooks which need metadata, covers, editing plus whatever else I want to do to modify them, the second library with (mine is called Main) gets the ebooks after they have been massaged. Main is where I send to my ereaders, etc.

And yes, the library names are rather obvious and boring.

When I was first moving to using calibre, I used some bulk metadata downloads but there were always problems with a goodly percentage of the ebooks so I now use individual metadata downloads. This takes a bit longer than the bulk download but less time overall since I don't have to redo the bad metadata downloads.

I do a similar thing. I have one library for my eBooks. I have another for the ones I plan to fix up and send to my Reader. The first library I leave the eBooks untouched. The other library is where I have the books I'll read after I've had a go at them.

[theducks]

The reasoning for a separate Intake Library, is you do not pollute the main database with (excessive) tags, poor naming of Author, publisher... .

Every discarded (field) entry is still in the DB, it is just hidden until you run Library Maintenance (the cleaning phase)