[smivan]

The challenge that I face is Calibre absolutely craters after 20,000+ ebooks.

I understand that most folks probably have 10-300 books, but in my case I have some large libraries I tried to use with Calibre, and it just absolutely DIES.

The steadfast decision to exclusively use SQLite doesn't help. I opened a PR back in 2016 or so, but the maintainer was quite enthusiastically opposed to changing the DB layer or making it pluggable.

[bomewish]

SQLite couldn’t be the problem. Scales well to millions and millions of rows. Clearly something else going on.

[Hendrikto]

> the maintainer was quite enthusiastically opposed to changing the DB layer or making it pluggable

Sounds very reasonable. I am pretty sure that SQLite isn‘t the problem. It is actually faster than traditional client server DBs in many cases.

[Onavo]

This is strange, Kovid Goyal, for all his apathy towards UI, is an excellent performance engineer. Check out his production terminal emulator kitty. He's more than capable of building extremely scalable systems. I would be surprised that this is a bottleneck caused by Calibre's core code.

He is however an asshole on the level of Linus when it comes to bug reports and community support.

[alsodumb]

This keeps getting parroted in every thread about Calibre, but Kovid is not an asshole when it comes to community support, it was more of a one time incident. Here is a comment I’m copying from a different HackerNews thread:

“ It's a big shame that whenever anything related to Calibre is posted, top comments will often be how arrogant is author is. I used to think the same. But when I read his comments history in GitHub, he doesn't seem a dick to me at all. So I googled his name, and end up in a forum. It completely changed my mind. Contrary to what people believe, Dovid Goyal is the most friendly programmer to his users. He spent lots of time answering questions in the forum https://www.mobileread.com/forums/forumdisplay.php?f=166. His answers include where is a menu located, how to convert a book from one format to another, helping the user to debug the exact the problem with Calibre. He has been doing this daily and is very responsive. You can expect an answer to your question in the same day. (The forum used to allow you to read user's comments history anonymously, but now you need a account to do so)”

[flkiwi]

That kitty and Calibre come from the same brain is a marvel of the open source world.

[quink]

Not really, kitty was forked from putty.

Calibre afaik was created from scratch.

Those are two entirely different circumstances from which to start from and it’s inevitable they would have vastly divergent results even in the same hands.

[aumerle]

No kitty is not a fork of putty. Thats a different kitty. https://github.com/kovidgoyal/kitty vs https://github.com/cyd01/KiTTY

[quink]

I stand corrected.

[dns_snek]

It wouldn't be a (sub-)thread about Calibre without someone pointing out the last bit. Made me chuckle.

[asymmetric]

What makes you think SQLite is the bottleneck?

[marvel_boy]

20.000 rows is a joke for SQLite (and for any database).

[nekopa]

What happens after 20K books? I am at about 13K books, and I haven't run into any issues yet. I make extensive use of virtual libraries in Calibre, and tags for organization.

[peekpeek]

Over 25,000 books. Still runs quite nicely even on a slower mini PC. I suspect you are using a unusual use case like keeping the library on a fat32 drive.

[retrochameleon]

This is one of the frustrations I have heard about Calibre's dev: very resistant to certain ideas or making Calibre more extensible.