[chaley]

Originally Posted by jecilop

@chaley,

???
Custom metadata may be embedded, but if you don't have the same custom columns in your database, then how do you retrieve it not knowing what to set up again?

If you don't know what custom columns you had then you are indeed in difficulty. That said, if you keep a copy of metadata_db_prefs_backup.json then you can see the definitions. You can also create an empty library as a backup, copying structure. This empty library can be used to recreate a library into which you import the books.

Also, it's more about being able to use any reader or app and utilize that info in some manner which I realize is a choice and/or a function of the scope of the app and not an issue with Calibre.

Your choice, to be sure. I never use the embedded metadata in a book so it is not something I think or care about.

For example, I foolishly think I can save all books to epub using Save to Disk inclusive of embedded data. I lose or delete ALL calibre data since I have my files saved via Save to Disk to another place. A get a new PC and decide I want the powerful library function back and add all books to a new, non-customized install of Calibre where no previous libraries and databases exist.
To my knowledge, the only thing I can retrieve from the epubs are the default metadata columnal info such at title and author and default tags, etc - especially because custom columns have been forgotten and thus not easily recreated from my memory... IS THIS WRONG?

Yes, it is wrong. Save the books in a single directory per book (done using the template) and include the metadata.opf file along with it. You can them import them into a calibre library using "add, one book per directory". All the metadata will be imported as long as the library has the correct definitions.

Of course, it is easier and faster to create a backup of the library by simply copying the library somewhere else. Doing so both preserves the library itself (as long as you keep the metadata.db file) and creates a directory structure that is easily imported.

Problem:
I looked over Search/Replace as mentioned. A previous post answer helped me understand it better. However, I cannot get it to work. I've attached a picture with a test example using 2 books and multiple tags. The goal is to get

• Book 1: Test A from {#tagsmgmt} to combine into {tags} that already has tags1, tags2; this would yield {tags} tags1, tags2, Test A
• Book2: Test B from {#tagsmgmt} to combine into {tags} that already has tags3, tags4; this would yield {tags} tags3, tags4, Test B

You need to use (.*) in the search expression and \1 in the replace expression. See the attachment.