BookSync-a web page to access Calibre Libraries on Dropbox

[slm]

BookSync is a rough-and-ready solution for accessing your Calibre Libraries stored on Dropbox from the internet. It is a simple web page, accessible from any device with an internet connection and a browser. (But it probably won’t work on most eReader browsers).

BookSync will not work correctly if your Calibre Libraries contain more than 10,000 books.

I am looking for beta testers. You can just go the website to use it: https://books.millmans.us/bluebooks.html If you click on a book title, you will get be a standard book details screen (Title, cover, author, and blurb) with a button to download the epub and/or azw formats, if available.

I wrote BookSync because of the problems people started having with Calibre Sync when Calibre modified its database structure. I was already using a personal program that looks like this one and—except for the way in which it accesses Dropbox--works the same way.

BookSync is a simple web page: it doesn’t collect any information, and it doesn’t need to be “installed” into iOS or Android to run. Although it reads files and folders in your DropBox account, it does not write to them.

Book Sync is fairly robust because it scans the Calibre folder structure, not the database and it uses standard Dropbox APIs to read file and folder data. It should be fairly low-maintenance, thus reducing the risk of becoming obsolete since it doesn’t have to keep up with the changes Apple and Google keep making to their app store requirements or changes Calibre may make to its database structure and it doesn’t need to be side-loaded. In addition, the source code is available to anyone who wants to run it (with necessary changes) from their own server.
It isn’t pretty—see screenshot. It takes about 2 seconds to build the list of my 3500 books. If you have a really big library, BookSync may not be for you. But it provides access from anywhere to the copy of your Calibre libraries you may have on your Dropbox account


A few non-features that don’t count as bugs:

1. The program shows all your Calibre books, not just the books from a single library. (In fact, it treats any folder that contains a file named “metadata.opf” as containing a book.)
2. The program gets the book title and author from the folder name. Since Calibre sometimes shortens book or author names when it creates folders (and sometimes changes some characters), this may not be the full title or author.
3. It doesn’t work with other cloud sync services: I know even less about these than I did about Dropbox. And I don’t have a Calibre library stored on any other cloud service, so I’m not going to try to extend the program to other cloud services. Any one else is welcome to do so, of course.

I am also looking for Beta Fixers. As mentioned, the source code is freely available (3 small files) and it can be run (with a few changes and some adjustments to the Dropbox console) on other servers. My general Javascript knowledge is middling and everything I know about authorizing Dropbox and accessing it through the API, I learned from Google Gemini for this project—some of it wrong. A lot of you can do a better job. I’m not a GitHub member, so send me a private message with an email address if you want the code (or just copy if from the Developer Tools).

I would also be happy to entertain ideas for a better name and an icon.

[slm]

Fixed a stupid choice, now scans all 3500 books in 2 seconds!
(I've changed some of the text in the first post)

[Sirtel]

Hmm. For some reason it only shows books from one of my libraries (the largest), but not from others. And filtering seems to be very buggy - when I try to filter by the name of the author, several duplicate entries are displayed and many entries aren't displayed.

[slm]

Thanks for trying this out.


I can replicate the filtering problem, but it isn't actually the filtering but the file list--for some reason, some books are "found" more than once in the search. This is "supposed" to happen, according to my very limited understanding.
I'm trying to figure out why is does.

As to missing your other library, a couple of questions may help:
1. how may books are in each library?
2. Can you open the folder for one of the books in the library that was missed and check that that folder does in fact contain a file whose name is "metadata.opf" (I can't imaging why it wouldn't but that is what the program searches for).
3. Finally, what OS are you on and what Browser? (Neither of these should matter, but it might be a place to start)

[Sirtel]

You mean metadata.db, I think? Yes, they all have it. And all the books have metadata.opf files as well.

The largest is over 35 000 books and alphabetically the first one. The other two (that the app couldn't find) are about 5000 and 3000.

And the issue with filtering is not only duplicates, it's also that many books for the filtered author(s) aren't shown at all. I think the app may have problems with huge libraries - maybe it's unable to process such a number of books and that's the reason for both missing libraries and missing books.

I'm on Windows 11, the browser is Thorium (a Chromium fork. It's the only one where I'm logged in to Dropbox).

[slm]

Originally Posted by Sirtel

You mean metadata.db, I think? Yes, they all have it. And all the books have metadata.opf files as well.

Actually, I really did mean "metadata.opf". Each library has a metadata.db file but each book has a metadata.opf file. (the "opf" file contains information, or references to the information that Calibre uses for the book. Calibre uses this to rebuild the database if it gets corrupted.)

[slm]

By the way, how long does it take to search 35,000 books?

[Sirtel]

Originally Posted by slm

By the way, how long does it take to search 35,000 books?

Less than a minute, but I suspect it's stopping long before it has scanned all of them. The list is displayed after it found 9000. So maybe that's the limit.

[slm]

Your guess turns out to be correct: I verified it empircally and then did a search of the documentation, which does state this limit.

I've modified the initial post to warn that the program does not work correctly with large libraries.

I've actually found the recommended approach for larger searches but (because I tried using this technique at one point), I think it will be too slow to be reasonable for large libraries. I'm going to play with it a bit more to be sure.

Thank you very much for your help.

[slm]

@sirtel: The duplicated files issue has been solved. Not that it will do you can good, of course, since the 10,000 books limit is still present.

[Sirtel]

Thanks.

It's a pity there's no option to select a library. I'd give it a go at one of my smaller ones.