How do I import this library ?

[Giuseppe Chillem]

Hello,
I have just downloaded a library of books which has this structure

Author Name\book name\file.mobi
Author Name\book name\cover.jpeg
Author Name\book name\metadata.opf

It has been created from a Calibre Library

Is there a way to add it to the calibre library folder and let calibre update his database ?

Thanks in advance

[DoctorOhh]

Originally Posted by Giuseppe Chillem

Hello,
I have just downloaded a library of books which has this structure

Author Name\book name\file.mobi
Author Name\book name\cover.jpeg
Author Name\book name\metadata.opf

You may find that it has a metadata.db file too. If it does just switch to this location and see what you have in the library. If you want to you can copy from this library to your current library.

Alternatively you could just add the entire directory of books to your collection using the add books from directories feature (see attached). If you do this you might want to have these two settings checked, see attached.

[Giuseppe Chillem]

Originally Posted by dwanthny

You may find that it has a metadata.db file too. If it does just switch to this location and see what you have in the library. If you want to you can copy from this library to your current library.

Alternatively you could just add the entire directory of books to your collection using the add books from directories feature (see attached). If you do this you might want to have these two settings checked, see attached.

Found the metadata.db and the copy library feature.

Everything seems working.

Thanks

[Mixx]

Can I please reopen this thread? In my case, regrettably, there is no *.db file. I just have a bunch of author_title.zip files with

book.html
book.jpg
book.opl
book.txt
mbp_toc.html

for each book, but put into a subdirectory called "title" in the ZIP file. When I import the ZIP file, the metadata are read properly, the cover is not loaded though and the format of the book is "ZIP". But when I try to read it or convert it, I get an error :TypeError: Invalid input object: NoneType.

Any help appreciated.

Mixx

[Manichean]

Originally Posted by Mixx

Can I please reopen this thread? In my case, regrettably, there is no *.db file. I just have a bunch of author_title.zip files with

book.html
book.jpg
book.opl
book.txt
mbp_toc.html

for each book, but put into a subdirectory called "title" in the ZIP file. When I import the ZIP file, the metadata are read properly, the cover is not loaded though and the format of the book is "ZIP". But when I try to read it or convert it, I get an error :TypeError: Invalid input object: NoneType

Try unzipping all the files, each file into it's own subdirectory (7zip allows that for multiple ZIPs with one click, don't know about other programs). Then, in Calibre, in the add books dropdown, select the option "subdirectories, one book per directory, multiple formats". If I'm not much mistaken, that ought to do it.

[Mixx]

Thanks, Manichean, for your suggestion.

In that case the metadata are read correctly again except for the Title (which is: "mbp toc" in each case), the cover is not read. There are two files in the entry: one TXT (with the small excerpt in the *.txt file), one ZIP, which can be displayed or converted, but it contains only (the correct) table of contents, with links (but these point to NIL).

Regards, Mixx

[Manichean]

First off, do you read metadata from file name or from file content (see preferences->adding books)? (I just realized that I confused OPL with OPF, I was thinking about Calibre reading the metadata from the OPF.)
What I find somewhat confusing is that you get only the TOC inside the ZIP file (Calibre zips up anything it thinks belongs to a HTML book, so the ZIP file part is expected behaviour), does the mbp_toc.html point to entries inside book.html? If not, what does it point to?

[Mixx]

Hi Manichean,

I read metadata from file content. And indeed, the ZIP file (cervantes.zip) contains these files:

META-INF
container.xml
content.opf
mbp_toc.html
mimetype

mbp_toc.html is small and contains only the TOC. The link all point into a TEMP directory of Windows c:\Users\MyID\AppData\Local\Temp\_tc\ in which an html file is created,

file:///C:/Users/MyName/AppData/Local/Temp/_tc/E%3A%5CBooks%5CE-Book%5Cdon_quijote%5Cdon_quijote.html#mbp_toc_9

Thus, the book itself is not there but one can see the title (Don Quijote) in the path above.

Thanks for your help,

Mixx

PS: container.xml is in subdir Meta-Inf

[Manichean]

Hm. I'm not entirely sure, but that file structure reminds me of the contents of an ePub file. Also, I'm somewhat confused, earlier you said the contents of the ZIP files were different?
Try renaming the files that have the content and structure as described in your last post to .epub and adding that file to Calibre.

[Mixx]

Appreciate your patience, Manichean!

The contents *are* different, but the file name of that one file "mbp_toc.html" is the same in all of them. When I just import any of the ZIP files, it pops up identical for each file.

I renamed the ZIP file into EPUB, as you suggested. I can import it all right, but the metadata are not read correctly. The file type is EPUB now, which I can not read with the built-in reader, but a conversion into MOBI is possible after which the file can be read.

With that I have a half-half situation: either I get the metadata right or the book, but not both. I'll see if I can import twice and merge.

The trouble is, even if that works, I'll still have to add the cover manually (grr) and will not have TOC (which is generated but the links point to nirvana).

Is there a way maybe, that I could produce "proper" epub files, since we are *this* close to them anyway?

Thanxx, Mixx

[Manichean]

Originally Posted by Mixx

I renamed the ZIP file into EPUB, as you suggested. I can import it all right, but the metadata are not read correctly. The file type is EPUB now, which I can not read with the built-in reader, but a conversion into MOBI is possible after which the file can be read.

With that I have a half-half situation: either I get the metadata right or the book, but not both. I'll see if I can import twice and merge.

I guess I was wrong then and those files are not falsely labelled ePub. Would've been too easy anyway.
Best solution I can think of right now is importing them as ePub anyway, since you said you were able to convert those files. Do an ePub->ePub conversion, that ought to give you valid files. Read the metadata from file name instead of file content while importing, and before converting, do a bulk metadata edit reading the cover from the file. That should get you nice metadata and readable books with the least amount of work, I believe.
Hint: Try this on a subset before committing all books

[Mixx]

Well, just to close this thread (or at least my part of it):

I ended up deleting the files

book.txt
mbp_toc.html

after which I could read the whole library, except for the book.jpg files (=covers).

book.txt is irrelevant and is not being referenced by the other files. mbp_toc.html is the table of contents and is being referenced in the main (book.html) file in the folder. This was apparently the problem, calibre can not include those files somehow, even though there is an explicit reference by filename in the main html file.

For me, mission accomplished, it is just that I have all those covers in the folders and I can not seem to find a way to read them together with the other files in the folder.

If I might, I'd like to suggest this as an improvement: a book.jpg file in the same folder as book.opf and book.html should be read together with the others and applied as the cover image. That would be handy in some cases.

Many thanks to Manichean for his help.

Thanxx, Mixx