[Idea]: Textbook Slicing

[jpazdjara]

What are you trying to do?

As a student, I use Open Notebook alongside my Obsidian vault, and I'd love to be able to upload my textbook PDFs directly. The challenge is that larger files come with high cache and token usage. I'm already working around this by generating summaries, but it would be fantastic if a textbook upload could be automatically split into chapters and grouped under a single textbook heading.

I'm not sure how high a priority this is, but it would be a really nice feature to have.

What feels difficult or missing today?

Right now, if I want to reference a textbook, I either have to manually split the PDF into smaller chunks before uploading, or rely on a separate tool to do it. At the moment I use a custom Obsidian plugin I built for this.

What outcome would help?

Being able to upload a textbook, see its chapters (I often only need a few), select which ones I want, and then upload just those to a notebook would be ideal.

How do you handle this today?

I built a custom Obsidian plugin that uploads my textbooks to Open Notebook as chapters, but it would be great to have this handled natively on the Open Notebook side.

Additional context, examples, or possible directions

Some mock ups of how I see it, if that helps. :)

How would you like to participate?

• I can add use cases or join the product/design discussion.
• I can help explore the technical approach.
• I can help implement an approved work item.
• I can test or validate a proposed solution.

Before posting

• I searched existing Discussions and Issues for related proposals.

[lfnovo]

Thanks — this is a model idea post: clear problem, a working workaround, mockups, and you ticked every participation box. Let me break down what I see, because there are two different design questions hiding in one feature:

1. Chapter detection/splitting — turning a textbook PDF into chapters at ingestion. Architecturally, this would most likely live in content-core, the extraction library Open Notebook uses for all source processing — detection there would benefit every consumer, and Open Notebook would build the selection/grouping experience on top. So: how does your Obsidian plugin detect chapters today (TOC, PDF bookmarks, heading heuristics)? How reliable is it across your textbooks? Real-world detection quality is exactly the input this exploration needs.
2. Grouping in the library — "chapters grouped under a single textbook heading" implies sources that belong to a parent. Today the source list is flat. Support for external/mounted sources, or source metadata/grouping? #291 has been asking for source grouping/metadata from another angle; whatever we design should answer both.
3. Selective inclusion — picking which chapters join a notebook. Follows naturally from (2); your mockups already sketch it.

One fork I'd like your take on: do you need the textbook preserved as one source with internal structure (chapters as sections you can select), or are independent chapter-sources visually grouped good enough? They look similar in the UI but are very different data models — and your plugin does the second today, so you have real experience with its limits.

Related: #1215 raises breaking huge documents into work units, and #1154 covers retrieval quality at scale — same gravitational field. Document-structure awareness aligns with where retrieval is heading, so this is a well-timed exploration.

Status: exploring — with a genuine offer: your plugin experience plus willingness to implement makes you a strong candidate to help shape and build this once the grouping model is decided.