New Calibre blazor web content server looking for Beta tester

[chris185]

Hello,

I have developed a new Calibre blazor web content server Calibre.Web.
It has the following features :
1. It maintains it's own postgre book database which is faster and is synced with your Calibre database
2. You can read books online.
3. You can convert selected books to other formats.
4. You can download selected books with the Calibe.Web download client for Windows or Linux. You can also import selected books directly in your local Calibre database.
5. You can upload or import EPUB books.
6. The App is multilingual in the final version in English and German until now in English. You can also add your own language and translate it automatically with DeepL or Bing translator API.
7. You can search in Google books or Internet Achive for ebooks and download them.
8. It is available for Windows and Linux.

The develpment is almost done and i am looking for people who want to Beta test it. So if you are interested in beta testing send me a email to: christophweller@dcm-works.net with info about the OS you are using.

Regards
Christoph Weller

[QuietShelfLife]

This looks like an interesting project -- nice to see someone building a modern web frontend for Calibre.

A few questions:

1. What made you go with a separate PostgreSQL database rather than reading Calibre's SQLite directly? I'd imagine keeping two databases in sync adds complexity, but maybe the query performance difference is significant for large libraries?

2. For the online reading feature -- are you rendering EPUBs in the browser? If so, are you using something like Readium.js or a custom reader? Browser-based EPUB rendering has some tricky edge cases with pagination and CSS support.

3. Does the Internet Archive integration use their Open Library API, or are you scraping search results? Their lending library API can be a bit finicky.

The DeepL/Bing translation integration for multilingual support is a clever touch. Would love to test this if you're still looking for testers.

[rantanplan]

Originally Posted by QuietShelfLife

1. What made you go with a separate PostgreSQL database rather than reading Calibre's SQLite directly? I'd imagine keeping two databases in sync adds complexity, but maybe the query performance difference is significant for large libraries?

Haven't you noticed all the threads about Calibre 9 breaking calibre-web and calibre-sync?

[chris185]

Originally Posted by QuietShelfLife

This looks like an interesting project -- nice to see someone building a modern web frontend for Calibre.

A few questions:

1. What made you go with a separate PostgreSQL database rather than reading Calibre's SQLite directly? I'd imagine keeping two databases in sync adds complexity, but maybe the query performance difference is significant for large libraries?

2. For the online reading feature -- are you rendering EPUBs in the browser? If so, are you using something like Readium.js or a custom reader? Browser-based EPUB rendering has some tricky edge cases with pagination and CSS support.

3. Does the Internet Archive integration use their Open Library API, or are you scraping search results? Their lending library API can be a bit finicky.

The DeepL/Bing translation integration for multilingual support is a clever touch. Would love to test this if you're still looking for testers.

Hello,

1. I have decided to use Postgresql because i need a place to store usermangement, languages and settings and other data. Also SqLite has a few limitations on multi-thread access.
2. Yes i display EPUB's in the browser without addiftional libs. I read the epub xml content convert it to html and strip off css-links and add images to the output. That will be displayed as a HTML markup string. That works fine.
3. I use their Open Library API.

And yes i am looking still for testers. Please send me an email to christophweller@dcm-works.net and tell me your OS and i give you the download-link. Thanks!

Best regards
Christoph Weller

[chris185]

Originally Posted by rantanplan

Haven't you noticed all the threads about Calibre 9 breaking calibre-web and calibre-sync?

Hello,

Thanks for the hint. I have made my App configurable so that you can choose the SQLite database version 9 or pre 9.

Best regards
Christoph Weller

[QuietShelfLife]

Thanks for the detailed answers, Christoph.

Originally Posted by mapwi

Haven't you noticed all the threads about Calibre 9 breaking calibre-web and calibre-sync?

Ha, fair point. The Calibre 9 schema changes caught a lot of third-party tools off guard. Having your own database with a sync layer means you're insulated from that -- you just update the import logic rather than having the whole app break when Calibre restructures its internals. And being configurable for pre-9 vs. 9 schemas is a nice touch.

Originally Posted by chris185

Yes i display EPUB's in the browser without additional libs. I read the epub xml content convert it to html and strip off css-links and add images to the output.

Rolling your own EPUB-to-HTML pipeline gives you full control, which is nice. One thing to watch out for: some EPUBs rely heavily on their CSS for basic readability -- things like text-indent for paragraph formatting, margin on blockquotes, or @font-face for embedded fonts. Stripping all CSS links can turn a well-formatted book into a wall of unstyled text. Might be worth selectively inlining the CSS (or at least the layout-affecting properties) rather than dropping it entirely.

Also, some EPUB3 books use epub:type attributes for semantic structure -- footnotes, endnotes, etc. Those won't mean anything to a browser, but they could be an opportunity to add nice pop-up footnote handling in the web UI.

I'll send you an email for the beta -- happy to throw some EPUBs with tricky formatting at it.

[chris185]

Originally Posted by QuietShelfLife

Thanks for the detailed answers, Christoph.



Ha, fair point. The Calibre 9 schema changes caught a lot of third-party tools off guard. Having your own database with a sync layer means you're insulated from that -- you just update the import logic rather than having the whole app break when Calibre restructures its internals. And being configurable for pre-9 vs. 9 schemas is a nice touch.



Rolling your own EPUB-to-HTML pipeline gives you full control, which is nice. One thing to watch out for: some EPUBs rely heavily on their CSS for basic readability -- things like text-indent for paragraph formatting, margin on blockquotes, or @font-face for embedded fonts. Stripping all CSS links can turn a well-formatted book into a wall of unstyled text. Might be worth selectively inlining the CSS (or at least the layout-affecting properties) rather than dropping it entirely.

Also, some EPUB3 books use epub:type attributes for semantic structure -- footnotes, endnotes, etc. Those won't mean anything to a browser, but they could be an opportunity to add nice pop-up footnote handling in the web UI.

I'll send you an email for the beta -- happy to throw some EPUBs with tricky formatting at it.

Hello,

Until now all epubs which i have tested where rendered correctly. Did you sent the email ? until now i have nothing received.

Best regards
Christoph Weller

[QuietShelfLife]

Sorry, got sidetracked and haven't had a chance yet. Will circle back when I do.

[TIHTCM]

Hi Christoph,

Hope you're doing well. I’ve been following your new project and it looks really promising, so I wanted to reach out. I’d definitely be interested in helping test it as things move forward — I run a pretty large Calibre setup and would love to put it through real‑world stress.

Two features I rely on heavily in Calibre are duplicate detection and full‑text indexing/search. Are those already implemented in your system, or planned for a later phase? They’re core parts of how I manage my library, so I’m curious how you’re handling (or planning to handle) those workflows in the new architecture.

I also had a technical question about the database model:
Is metadata.db still required, or is your PostgreSQL database intended to be the primary/only metadata store?

My library is over 50k books, and the SQLite metadata.db tends to corrupt more often than I’d like. If your project allows relying on PostgreSQL alone, that would solve a major stability issue for me.

Thanks again for putting this together — happy to help test, stress, and break things if that’s useful.

Best,

TM
https://www.mobileread.com/forums/sh...54#post4575254

Originally Posted by chris185

Hello,

I have developed a new Calibre blazor web content server Calibre.Web.
It has the following features :
1. It maintains it's own postgre book database which is faster and is synced with your Calibre database
2. You can read books online.
3. You can convert selected books to other formats.
4. You can download selected books with the Calibe.Web download client for Windows or Linux. You can also import selected books directly in your local Calibre database.
5. You can upload or import EPUB books.
6. The App is multilingual in the final version in English and German until now in English. You can also add your own language and translate it automatically with DeepL or Bing translator API.
7. You can search in Google books or Internet Achive for ebooks and download them.
8. It is available for Windows and Linux.

The develpment is almost done and i am looking for people who want to Beta test it. So if you are interested in beta testing send me a email to: christophweller@dcm-works.net with info about the OS you are using.

Regards
Christoph Weller

[chris185]

Originally Posted by TIHTCM

Hi Christoph,

Hope you're doing well. I’ve been following your new project and it looks really promising, so I wanted to reach out. I’d definitely be interested in helping test it as things move forward — I run a pretty large Calibre setup and would love to put it through real‑world stress.

Two features I rely on heavily in Calibre are duplicate detection and full‑text indexing/search. Are those already implemented in your system, or planned for a later phase? They’re core parts of how I manage my library, so I’m curious how you’re handling (or planning to handle) those workflows in the new architecture.

I also had a technical question about the database model:
Is metadata.db still required, or is your PostgreSQL database intended to be the primary/only metadata store?

My library is over 50k books, and the SQLite metadata.db tends to corrupt more often than I’d like. If your project allows relying on PostgreSQL alone, that would solve a major stability issue for me.

Thanks again for putting this together — happy to help test, stress, and break things if that’s useful.

Best,

TM
https://www.mobileread.com/forums/sh...54#post4575254

Hi TM,

Thanks for your mail. Testing is already done and the App is just released, see my other forum message. You can download it from https://www.dcm-works.net/

Regarding your questions:
1. The App detects duplicates during import process.
2. There is a full text index search on the books page. You can also filter on the categories.
3. The App still relies on the SQLite metadata.db. But i think it would be easy to implement that you can switch it off and working only on the postgresql database. Please check out the App and if you still have problems with the SQLite database return to me and i will implement it. Thanks!

Best regards
Christoph Weller