MobileRead Forums - View Single Post - Question about accessing custom column information

wavioto · 03-13-2025, 09:16 AM

I've created a custom column in Calibre to store the contents of KOReader sidecar files (specifically, the notes generated while reading). I'd like to access the text from these notes and export them to Obsidian.

My initial attempt was to use Python and sqlite3 to directly access the Calibre .calnotes/notes.db file. However, I encountered errors, and after some research, I discovered that Calibre uses a custom tokenizer, which complicates direct SQL queries. I'm also not sure if notes.db is the correct database to be looking at for what I am attempting to do.

I'm a beginner in programming in general and Calibre internals, so I'm a bit lost.

Is .calnotes/notes.db the correct database?
If yes, what do I do for the custom tokenizer?

Is my approach valid or should I change it?

Any help and pointers would be greatly appreciated.

03-13-2025, 09:16 AM	#1
wavioto Junior Member Posts: 8 Karma: 10 Join Date: Mar 2025 Device: KT4 (kindle 10th generation)	Question about accessing custom column information I've created a custom column in Calibre to store the contents of KOReader sidecar files (specifically, the notes generated while reading). I'd like to access the text from these notes and export them to Obsidian. My initial attempt was to use Python and sqlite3 to directly access the Calibre .calnotes/notes.db file. However, I encountered errors, and after some research, I discovered that Calibre uses a custom tokenizer, which complicates direct SQL queries. I'm also not sure if notes.db is the correct database to be looking at for what I am attempting to do. I'm a beginner in programming in general and Calibre internals, so I'm a bit lost. Is .calnotes/notes.db the correct database? If yes, what do I do for the custom tokenizer? Is my approach valid or should I change it? Any help and pointers would be greatly appreciated.